Scoulerine affects microtubule structure, inhibits proliferation, arrests cell cycle and thus culminates in the apoptotic death of cancer cells

Scoulerine is an isoquinoline alkaloid, which indicated promising suppression of cancer cells growth. However, the mode of action (MOA) remained unclear. Cytotoxic and antiproliferative properties were determined in this study. Scoulerine reduces the mitochondrial dehydrogenases activity of the evaluated leukemic cells with IC50 values ranging from 2.7 to 6.5 µM. The xCELLigence system revealed that scoulerine exerted potent antiproliferative activity in lung, ovarian and breast carcinoma cell lines. Jurkat and MOLT-4 leukemic cells treated with scoulerine were decreased in proliferation and viability. Scoulerine acted to inhibit proliferation through inducing G2 or M-phase cell cycle arrest, which correlates well with the observed breakdown of the microtubule network, increased Chk1 Ser345, Chk2 Thr68 and mitotic H3 Ser10 phosphorylation. Scoulerine was able to activate apoptosis, as determined by p53 upregulation, increase caspase activity, Annexin V and TUNEL labeling. Results highlight the potent antiproliferative and proapoptotic function of scoulerine in cancer cells caused by its ability to interfere with the microtubule elements of the cytoskeleton, checkpoint kinase signaling and p53 proteins. This is the first study of the mechanism of scoulerine at cellular and molecular level. Scoulerine is a potent antimitotic compound and that it merits further investigation as an anticancer drug.

SCiEntiFiC RepoRts | (2018) 8:4829 | DOI: 10.1038/s41598-018-22862-0 made from reticuline and serves as the precursor in the biosynthetic pathway for berberine, stylopine, protopine and sanginarine 6,7 . An initial biological study described its significant in vitro antiplasmodial activity against the P. falciparum strains, TM4/8.2 (a wild type chloroquine and antifolate sensitive strain) and K1CB1 (multidrug resistant strain), with IC 50 values 1.78 µg/mL and 1.04 µg/mL, respectively. Regrettably, this activity does not meet the criteria stipulated under the Medicines for Malaria Venture 3 . Other research efforts, performed on rats, determined that scoulerine protects α-adrenoreceptors against irreversible blockade by phenoxybenzamine, inhibits [ 3 H]-inositol monophosphate formation caused by noradrenaline 8 and acts as a selective α 1D -adrenoreceptor antagonist without affecting the contraction of the rat aorta 9 . Scoulerine has also been reported to exhibit other useful pharmacological properties such as antiemetic, antitussive and antibacterial action 3 and has been found to have an affinity to the GABA receptors 2 . Interestingly, a pioneer cell culture study on this alkaloid described that scoulerine shows significant cytotoxic activity against A549 and HT-29 cancer cell lines. The authors imply that the cytotoxic potency of scoulerine is associated with its ability to stabilize the covalent topoisomerase I -DNA complex to promote the formation of single-strand DNA breaks 10 . It should be pointed out that the unique position of scoulerine in backbone arrangements during biosynthesis and its interesting biological activities already attracted our attention in two previous studies. Scoulerine was found to be active as an inhibitor of ß-site amyloid precursor protein cleaving enzyme 1 (BACE1), which is a very promising target for the treatment of Alzheimer's disease (AD) 5 . In our follow-up work, when considering forty-six isoquinoline alkaloids screened by MTT assay, scoulerine exhibited impressive cytostatic activity against gastrointestinal cancer cells 11 . Although our recent study demonstrated the bioactivity of scoulerine with an emphasis on the cytostatic action that may be of interest in cancer chemotherapy, further studies remain to be undertaken to better explore its anticancer potential. At present, this study provides a better investigation of the MOA of scoulerine at cellular and molecular level. In addition to that, the pro-apoptotic and cell cycle arrest activity in p53-deficient (Jurkat) and p53 wild-type (MOLT-4) leukemic cells following scoulerine treatment is determined. Finally, aiming at the further conceptual extension to study structure-cytotoxicity relationships, we have introduced three (2, 3 and 4) aliphatic derivates of scoulerine incorporating esters of carboxylic acids.

Scoulerine decreases proliferation of human cancer cells. Various leukemic cell lines and carcinoma
cell lines obtained from solid tumors were used for evaluating their sensitivity towards scoulerine. First, scoulerine (1) and its three semi-synthetic derivatives (2), (3) and (4) were evaluated at a single-dose exposure (10 µM) against a mini-panel of leukemic cell lines (MOLT-4, Jurkat, Raji, HL-60, U-937 and HEL 92.1.7) using the XTT metabolic activity assay ( Fig. 2A) and A2780 ovarian carcinoma cells using the xCELLigence system (Fig. 2B). In order to determine the IC 50 values, a broad concentration range of scoulerine were determined ( Supplementary  Fig. 1). Considering the effect of aliphatic esters of scoulerine with varying carbon chain lengths, parent scoulerine proved to be the most active compound against the determined cell lines, with the IC 50 value ranging from 2.7 µM to 6.5 µM in human leukemic cells (Table 1). Further testing of the scoulerine effects on viability and proliferation revealed that in concentrations of 2.5, 5, 10, 15 and 20 µM scoulerine significantly reduced the viability and proliferation of Jurkat and MOLT-4 cells in a dose dependent manner within 24 h of treatment. Moreover, the reduction of cell viability was even more pronounced 48 h after scoulerine application ( Supplementary Fig. 2). Similar results were achieved using the dynamic real-time monitoring of proliferation by the xCELLigence system designated for adherent cell lines. In these experiments, the xCELLigence system was utilized to follow the proliferation of scoulerine and sham-treated cells over a 72 h period. As shown in Fig. 3, scoulerine (10, 20 and 50 µM) had a clear negative impact on human lung carcinoma (A549), ovarian carcinoma (A2780) and breast adenocarcinoma (SK-BR-3 and MCF-7) proliferation. Cells treated with lower doses (1 and 5 µM) of scoulerine experienced unaffected or slightly delayed proliferation.

Scoulerine induces MOLT-4 and Jurkat cells apoptosis.
In what concerns the pro-apoptotic activity of scoulerine in both Jurkat and MOLT-4 leukemic cell lines, this compound showed an apoptosis-inducing effect assayed by Annexin V and PI staining 24 h following the treatment. Soon after initiating apoptosis, the membrane phosphatidylserine was translocated from the inner layer of the plasma membrane to the outer leaflet. Quantification of Annexin V binding to externalized phosphatidylserine allowed us to clearly identify apoptotic cells between cell populations. Twenty-four hours after the application of 0, 2.5, 5, 10, 15 and 20 µM of scoulerine, the early apoptotic rates were 4%, 8%, 22%, 24%, 22% and 23% for Jurkat cells, and 3%, 4%, 9%, 16%, 14% and 13% for MOLT-4 cells, respectively. The late apoptotic (cells that have lost their membrane integrity also show PI staining) rates for Jurkat leukemic cells were 5%, 15%, 21%, 21%, 19% and 22%, respectively, and 5%, 14%, 20%, 27%, 26% and 28% for MOLT-4, respectively (Fig. 4). In follow-up experiments, the DNA fragmentation, which is a hallmark of apoptosis due to the activation of endonucleases, was quantified by TUNEL assay. In contrast Data are shown as mean values ± SD of at least three independent experiments. *Significantly different to control (P ≤ 0.05) (A). Dynamic real-time xCELLigence screen of proliferation and cytotoxicity over 62 h. The human A2780 ovarian carcinoma cells in the logarithmic growth phase were treated. The negative control cells were exposed to 0.1% DMSO (vehicle) and 5% DMSO was used as a positive control. The plot is representative of at least three experiments performed (B).  Scoulerine activates caspase-3/7, -8 and -9 in a dose-dependent manner. In order to subsequently confirm apoptosis activation at a molecular level, we evaluated the activation of caspases-3/7, -8 and -9. Irrespective of which pathway of apoptosis is induced, caspases-3/7 are activated downstream of initiator caspases. Caspase-9 is activated upstream of caspases-3/7 and downstream of mitochondrial membrane permeabilization, so it indicates that the intrinsic pathway of apoptosis is being activated, while caspase-8 becomes induced after death-receptor stimulation, which is typical for extrinsically mediated apoptosis. In this case, exposure of Jurkat cells for 24 h to 2.5 and 5 µM of scoulerine resulted in considerable (P ≤ 0.05) activation of caspases-3/7 and -8, and a bit less pronounced (P ≤ 0.05) activation of caspase-9. At the 48 h interval, the activity of caspases-3/7, -8 and -9 was still statistically significant, but began to decline. On the contrary, where MOLT-4 leukemic cells were concerned, the activity of caspases-3/7, -8 and -9 was even more pronounced 48 h following the application of 2.5 and 5 µM of scoulerine, compared to the same after 24 h, with an overall higher level of activity of caspases-3/7, -8 and -9 (Fig. 5).
Scoulerine acts to inhibit proliferation through inducing G2 or M cell cycle arrest. To further elucidate the mechanisms by which scoulerine exerts its growth-inhibitory activity, we investigated the effect of scoulerine on cell cycle progression in Jurkat and MOLT-4 leukemic cells 16 h after treatment. In the presence of 5 µM of scoulerine, the percentage of Jurkat cells in the G1phase and the S phase decreased from 45% and 31% (at mock-treated control) to 29% and 22%, respectively. Conversely, the proportion of cells in the G2/M phase increased from 24% (mock-treated control) to 49%. The rise in the percentage of Jurkat cells in the G2/M phase was even more eminent in the presence of 10, 15 and 20 µM of scoulerine. A similar trend was observed in MOLT-4 cells treated with the same amounts of scoulerine (Fig. 6). All in all, this result indicates scoulerine-induced cell cycle arrest at the G2/M transition, which was determined in later experiments by phosphorylated histone H3 at Ser10 (H3 pSer10) quantification. Analysis of the percentage of cells halting proliferation in mitosis (M phase) measured as positive for H3 pSer10 staining using flow cytometry revealed that scoulerine-treated Jurkat (2.5-20 µM) and MOLT-4 (5-20 µM) cells for 24 h were significantly (P ≤ 0.05) arrested in mitosis. Together, these results indicate that scoulerine is able to induce cell cycle arrest in the G2 or M phase depending on the dose and the duration of the treatment (Fig. 7).
Scoulerine disrupts microtubule structure of A549 lung carcinoma cells. Since we observed significant G2 or M phase arrest and H3 pSer10 staining during flow cytometry cell cycle analysis, we explored whether scoulerine affects the microtubule structures of cells. To visualize this, lung carcinoma A549 cells were chosen as a model for indirect immunofluorescence with a monoclonal anti-β-tubulin antibody due to a lower nuclear-cytoplasmic ratio compared to leukemic cells. Additionally, A549 cells were sensitive to scoulerine and displayed strongly diminished proliferation during xCELLigence analysis at concentrations exceeding 10 µM. Epi-fluorescence microscopy performed with fixed A549 cells labeled with a monoclonal anti-β-tubulin antibody and DAPI counter stain revealed that the intact microtubules extended continuously through the cytoplasm in control cells. In contrast, scoulerine applied for 24 h at 10 and 20 µM obviously disrupted microtubule structure with a dense aggregation of microtubules around cell nuclei, being less dense at the cell periphery. Besides, A549 cells treated with scoulerine at 5 µM resisted the treatment with almost intact organized tubulin networks as seen in 0.1% DMSO exposed control cells. Nocodazole, the microtubule depolymerizer, was used as a positive control for the study. As is apparent from the microscopic images presented in Fig. 8, treatment with nocodazole at 5 µM resulted in the observation of expected disorganized and disassembled microtubule structures as compared to 0.1% DMSO control cells.

Scoulerine does not induce DNA strand breaks.
To assess DNA damage, the alkaline comet assay was performed. The treatment of Jurkat and MOLT-4 cells with scoulerine at 2.5 and 5 µM for 24 h, triggered apoptotic (secondary) DNA fragmentation, which was estimated by the formation of comet-like tails after single-cell gel electrophoresis. Briefly, in a 24 h interval, there was a significant increase in the tail moment after scoulerine application in both Jurkat and MOLT-4 leukemic cells. However, when we performed the comet assay at an earlier 12 h interval of treatment, no considerable direct DNA damage was found. This suggests that DNA damage is predominantly a consequence of internucleosomal cleavage associated with apoptotic cell death in response to scoulerine treatment (Fig. 9).

Scoulerine alters the levels of cell cycle checkpoint and apoptosis-related proteins. In order
to study the molecular mechanism by which scoulerine induces cell cycle arrest and apoptosis, we investigated the expression levels and the activation of several cell cycle regulatory and proapoptotic proteins. To perform this, Jurkat and MOLT-4 cells were treated with scoulerine at 2.5 and 5 µM for 24 h and whole cell lysates were prepared and used for Western blotting. Western blot analysis showed an upregulation of p53 protein in p53 wild-type MOLT-4 cells in the presence of 5 µM scoulerine. In Jurkat cells, checkpoint kinase 1 (Chk1) was activated through phosphorylation at Ser345 and checkpoint kinase 2 (Chk2) was activated through phosphorylation at Thr68 after scoulerine treatment. However, in both determined cell lines the expression levels of p21, which is also involved in the regulation of cell cycle progression, remained the same. Similarly, the level of phosphorylated p53 at Ser15, which increased after cisplatin exposure (positive control), remained unchanged after scoulerine exposure (Fig. 10).

Discussion
In our previous study we reported that scoulerine has a potent antiproliferative activity against Caco-2 and Hep-G2 cancer cells 11 . Another recent work in this direction showed that scoulerine isolated from the stems of   Xylopia laevigata was cytotoxic toward the tumor cell lines B16-F10, HepG2, K562 and HL-60 12 . Encouraging results prompted us to investigate whether scoulerine can eliminate cancer cells via apoptosis and if the scoulerine-induced antiproliferative effect blocks cell cycle progression. Thus, in the work herein, we have investigated proliferation, cell cycle distribution, cell death, apoptosis induction, DNA damage, microtubule structure and the upregulation of selected DNA-damage response proteins following scoulerine treatment. We show that scoulerine had cytostatic activity in all of the leukemic and tumor lines investigated in a dose-dependent manner. Controversially, our results are in contrast with that reported by Khamis and colleagues. They determined only moderate cytotoxic activity of discretamine (scoulerine) with IC 50 over 3000 µM using four human breast cancer (MCF-7, MCF-7ADR, MDA-MB435 and MT-1) cell lines and MTT assay 13 . Here, however, scoulerine inhibited the proliferation of MCF-7 cells at 10 µM, as measured by means of the xCELLigence system in view of the cell-growth inhibition profile under real-time. To better understand antiproliferative potential of this naturally occurring alkaloid, derivatives of scoulerine were synthesized and assayed at 10 µM with respect to its activity on cell growth. The comparison of the semisynthetic derivatives (2), (3) and (4) with scoulerine indicated that esterification generally resulted in a reduction in the antiproliferative activity. Moreover, it seems that the potencies of the (2), (3) and (4) esters decreased with an increasing length of the carbon chain. Next, we observed that scoulerine treatment significantly downregulated the viability of Jurkat and MOLT-4 cells within 48 h from 5 μM and higher as assessed by using a Trypan blue exclusion test. Apoptosis induction associated with the translocation of phosphatidylserine to the outside of the cell and loss of plasma membrane integrity was quantified by flow cytometry. Jurkat and MOLT-4 cell cultures treated with scoulerine showed a presence of both early and late apoptotic cells, which increased in a dose-dependent manner. Apoptosis induction was further supported by measuring DNA fragmentation using TUNEL assay, by quantifying caspase activity and by detecting p53 expression in MOLT-4 cells. Collectively, scoulerine showed caspases-3/7, -8 and -9 activation, which suggests the involvement of both an extrinsic and intrinsic pathway of programmed cell death. Further analysis by flow cytometry revealed that scoulerine treatment led to G2 phase arrest in Jurkat and MOLT-4 leukemic cells followed by an increase in the percentage of cell population in mitosis as judged by PI and phospho-histone H3 double staining. During mitotic division, the tubulin-microtubule protein system underwent assembly, which, collectively, is an essential component for chromosome segregation. Where antimitotic compounds are concerned, this protein system is a suitable target. It is these antimitotic compounds that interfere with the microtubules function by inhibiting cell proliferation in mitosis 14 . Since uncontrolled and rapid cell division is a hallmark of cancer tumor growth, microtubule-interfering agents display a remarkable efficacy in rapidly proliferating cancer cells 15 . Therefore, we sought to elucidate, using immunofluorescence imaging, whether treatment with scoulerine affects the cytoskeletal microtubule architecture of cells. The epi-fluorescence images obtained, where A549 cells were exposed to 10 and 20 μM of scoulerine for 24 h and stained with an anti-β-tubulin antibody, revealed a disruption of the tubulin structure. These results support the notion that scoulerine is a mitotic poison, resulting in G2 or M arrest. Unlike drugs grouped as genotoxic anticancer agents, our results indicate that scoulerine does not interfere with DNA, thereby activating cancer cell death independently of exogenous-induced DNA damage. However, these findings oppose what has been reported by Cheng, who proposed that scoulerine's anticancer activity is associated with the formation of single-strand DNA breaks through the stabilization of the covalent topoisomerase I -DNA complex 10 . Thus, dissimilarly to the previous findings describing the DNA-damaging action of scoulerine, no significant DNA damage was found early at 12 h using the alkaline comet assay. However, as we expected, the later 24 h interval of treatment induced DNA fragmentation that resulted from an internucleosomal cleavage of DNA during scoulerine-induced apoptosis. Such single-or double-strand DNA breaks generated during apoptotic cell death were cross-verified by TUNEL at 24 h in both leukemic cell lines. Numerous reports have shown that replication stress and DNA damage activate critical cell-cycle checkpoint kinases Chk1 and Chk2 of the DNA damage response pathway 16,17 . Both Chk1 phosphorylated at Ser345 by Ataxia telangiectasia and Rad3-related (ATR) kinase and Chk2 phosphorylated at Thr68 by Ataxia telangiectasia mutated (ATM) kinase were determined after 24 h of treatment. These results imply that activated Chk1 and Chk2 are particularly important to pause cell cycle progression after scoulerine exposure.
Although scoulerine has been studied for several years, only few publications regarding its cytotoxic activity against mammalian cells exist. The most explored part is its biosynthesis in the plants of the  Nocodazole, an antineoplastic agent that disrupts microtubule function by binding to tubulin was used as a reference compound in this assay. Scale bar: 10 µm. Experiments were performed in triplicate using epifluorescence microscopy. Photographs from representative chambers are shown. Compared with controls, thicker and denser microtubule bundles were evident in scoulerine-treated cells.  subsequent benzo[c]phenanthridine, protopine and protoberberine alkaloids, sanguinarine, protopine and berberine, respectively 6,18 . Among these, it is important to point out that sanguinarine, protopine and berberine were reported to show various degrees of cytotoxicity and antiproliferative activity against cancer cells. Although sanguinarine did not change the proportion of mitotic cells, it inhibited tubulin polymerization leading to the disruption of the microtubule network 19 . Interestingly, protopine is able to target microtubule structures in living cells without affecting tubulin polymerization with cell cycle arrest at G2/M 20 . In the same manner, disruptions of the microtubule network were described in response to berberine treatment 21 . Even if this structurally related substance shares some mechanistic similarities, it should be noted that these compounds differ in growth inhibitory values and molecular actions, thus making them an attractive target for structure-activity relationship studies.
In summary, scoulerine was shown to be broadly cytostatic and cytotoxic in the range of micromolar concentrations. To further investigate the underlying MOA, we showed that scoulerine caused a pronounced accumulation of cells in the G2 or M phases of the cell cycle, increased phosphorylation of histone H3 at Ser10 and disruption of microtubule organization, which is consistent with the antimitotic mechanism of action. Additionally, scoulerine has shown molecular target activity that occurs during apoptosis, with a p53 protein increase, caspases-3/7, -8 and -9 activation, phosphatidylserine externalization and DNA fragmentation. Our findings suggested that scoulerine activated ATR and ATM kinase-dependent cell cycle checkpoint signaling followed by phosphorylation of Chk1 at Ser345 and Chk2 at Thr68 sites. To our best knowledge, this is the first example to show the potent antimitotic activity of scoulerine leading to microtubule disruption, cell cycle arrest and cell death via apoptosis induction in cancer cells. Altogether, our results identified isoquinoline alkaloid scoulerine as a potent microtubule targeting agent, and data on biochemical interactions playing an important role in the cytotoxic, antiproliferative and proapoptotic actions are sufficiently encouraging to warrant further research on its anticancer potential. General procedure for acylation of scoulerine. The 1.5 equiv. of the corresponding anhydride was added to a solution of scoulerine (1) in 3 mL of dry pyridine. The mixture was stirred at room temperature until disappearance of the starting material. Then, the solvent was evaporated and the residue was purified by preparative TLC using cHx:Et 2 NH 9:1 or cHx:To:Et 2 NH 45:45:10 to afford corresponding esters 2-4 22 .   149.2, 149.0, 138.1, 136.1, 132.9, 130.0, 128.1, 127.5, 126.3, 119.7, 112.2, 110.5, 58.6, 56.0, 55.9, 53.3, 51.1, 35.8,  29.4, 27.3, 27.2, 9.3
The purity of all compounds verified by NMR was ≥97%. Cytotoxicity screening using XTT assay. In order to determine cell viability of cells treated with scoulerine, alkaloid esters (a single-dose of 10 µM) or scoulerine in a broad concentration range we used a standard colorimetric method measuring a tetrazolium salt reduction via mitochondrial dehydrogenase activity. The cells were seeded at previously established optimal density in a 96-well plate. After 48 hours incubation cell viability was determined using Cell Proliferation Kit II (XTT, Roche, Germany) according to manufacturer's instructions. XTT-assay was conducted using 200 μl of volume and 100 μl of XTT-labeling mixture. Absorbance was then measured at 480 nm using a 96-multiwell microplate reader Tecan Infinite M200 (Tecan Group Ltd., Männedorf, Switzerland). Viability was calculated as described in the paper by Havelek and colleagues using the following formula: (%) viability = (A480sample − A480blank)/(A480control − A480blank) × 100, where A480 is the absorbance of utilized XTT formazan measured at 480 nm 23 . Data were analysed with GraphPad Prism 5 biostatistics (GraphPad Software, La Jolla, CA, USA) statistical software. Each value is the mean of at least three independent replicates of each condition.
Screening for antiproliferative activity using xCELLigence system. The  Single Plate station was placed inside the incubator at 37 °C and 5% CO 2 . First, the optimal seeding concentration for experiments was optimized for each cell line. After seeding, the respective number of cells in 190 µL medium per each well of the E-plate 96, the proliferation, attachment and spreading of the cells were monitored every 30 minutes by the xCELLigence system. Approximately 24 h after seeding, when the cells were in the log growth phase, the cells were exposed in triplicates to 10 µL sterile deionized water containing scoulerine to obtain final concentrations 1-50 μM. Controls received sterile deionized water + DMSO with a final concentration of 0.1%. Cells treated with 5% DMSO were used as positive control. Growth curves were normalized to the time point of treatment. Evaluations were performed using xCELLigence 1.

Activity of caspases.
The induction of programmed cell death was determined by monitoring the activities of caspases 3/7, caspase 8 and caspase 9 by Caspase-Glo Assays (Promega, Madison, WI, USA) 24 and 48 h after treatment with 2.5 and 5 μM of scoulerine. Cells treated with 5 µM cisplatin were used as positive control. The assay provides a proluminogenic substrate in an optimized buffer system. The addition of a Caspase-Glo Reagent results in cell lysis, followed by caspase cleavage of the substrate and the generation of a luminescent signal. A total of 1 × 10 4 cells were seeded per well using a 96-well-plate format (Sigma-Aldrich, St. Louis, MO, USA). After treatment, the Caspase-Glo Assay Reagent was added to each well (50 μl/well) and incubated for 30 minutes before luminescence was measured using a Tecan Infinite M200 spectrometer (Tecan Group, Männedorf, Switzerland). Epi-fluorescence microscopy. For each condition, 250 000 cells were seeded in 2-well chamber slides SPL (SPL Life Sciences, Korea). After seeding (usually 24 h later), spent medium was replaced with fresh medium and the cells were treated with scoulerine at 5, 10 and 20 µM. Cells treated with 5 µM nocodazole were used as positive control. Following 24-h treatment the cells were fixed with 4% freshly prepared paraformaldehyde for 10 minutes