Effects of Sepantronium Bromide (YM-155) on the Whole Transcriptome of MDA-MB-231 Cells: Highlight on Impaired ATR/ATM Fanconi Anemia DNA Damage Response

Abstract

Sepantronium bromide (YM-155) is believed to elicit apoptosis and mitotic arrest in tumor cells by reducing (BIRC5, survivin) mRNA. In this study, we monitored changes in survivin mRNA and protein after treating MDA-MB-231 cells with YM-155 concurrent with evaluation of whole transcriptomic (WT) mRNA and long intergenic non-coding RNA at 2 time points: 8 h sub-lethal (83 ng/mL) and 20 h at the LC₅₀ (14.6 ng/mL). The data show a tight association between cell death and the precipitating loss of survivin protein and mRNA (-2.67 fold-change (FC), p<0.001) at 20 h, questioning if the decline in survivin is attributed to cell death or drug impact. The meager loss of survivin mRNA was overshadowed by enormous differential change to the WT in both magnitude and significance for over 2000 differentially up/down-regulated transcripts: (+22 FC to -12 FC, p<0.001). The data show YM-155 to up-regulate transcripts in control of circadian rhythm (NOCT, PER, BHLHe40, NFIL3), tumor suppression (SIK1, FOSB), histone methylation (KDM6B) and negative feedback of NF-kappa B signaling (TNFAIP3). Down-regulated transcripts by YM-155 include glucuronidase (GUSBP3), numerous micro-RNAs, DNA damage repair elements (CENPI, POLQ, RAD54B) and the most affected system was the ataxia-telangiectasia mutated (ATM)/Fanconi anemia E3 monoubiquitin ligase core complexes (FANC transcripts - A/B/E/F/G/M), FANC2, FANCI, BRCA1, BRCA2, RAD51, PALB2 gene and ATR (ATM- and Rad3-Related) pathway. In conclusion, these findings suggest that a primary target of YM-155 is the loss of replicative DNA repair systems.

Breast cancer is one of the most common malignancies in females worldwide. Survivin (encoded by the BIRC5 gene) is classified as an inhibitor of apoptosis protein (IAP) which is overexpressed in various malignancies (1-5), when compared to adjacent non-cancerous tissue (6,7). Overexpression of survivin correlates to lack of tumor senescence and unbridled mitosis as a result of binding to the inner centromere protein (ICP), aurora B kinase and borealin to form the chromosomal passenger complex (CPC), required for prophase and metaphase events (8,9). The overexpression of survivin in malignant breast cancer models is associated with aggressive tumor growth, poor disease-free survival period, greater relapse (10,11) and taxane-based chemotherapeutic resistance, suggesting its role in triple negative breast cancers (TNBC) could be pivotal (12). Briefly, TNBCs are a sub-type of breast cancer characterized by the absence of estrogen and progesterone receptors and HER2/neu. This type of breast cancer has an overall poor prognosis worsened by the non-feasibility of hormone receptor-based treatment options, rendering the reliance on taxane-based chemotherapies.

One of the original survivin inhibitors identified was sepantronium bromide (YM-155), a first-in-class mRNA BIRC5 inhibitor showing therapeutic potential to antagonize tumor growth, induce apoptosis and decrease mitotic indices (13). The broad and potent anti-tumor activity of YM-155 has been reported against a wide variety of human cancer cell lines and xenograft models (14) where many cell lines have been found to be exquisitely sensitive to its lethal effects in the nM range, including TNBCs (15). Previous studies evaluating the effects of YM-155 in tumor models demonstrate capacity to halt cell proliferation, reverse chemo-resistance of statin based drugs (16) and in vivo establish the regression of subcutaneously implanted tumors including the MDA-MB-231 cell line in an orthotopic model (14). Agents that suppress survivin protein levels or knock out the gene appear to induce similar effects in both slowing tumor growth and providing greater sensitization to drugs such as gemcitabine (17-20) .

Although YM-155 has been described as one of the first survivin antagonists, its molecular mechanism still remains obscure. Moreover, there is a lack of public database microarray deposition for this compound in tumor models, which would define off target effects of this drug. In this study, we evaluate the effects of YM-155 on the entire transcriptome of MDA-MB-231 cells using Affymetrix WT human 2.1 ST Arrays. The data discussed in this publication have been deposited in NCBI’s Gene Expression Omnibus and are accessible through GEO Series accession number GSE103089 for public analysis.

Materials and Methods

Hanks Balanced Salt Solution, 96 well plates, pipette tips, general reagents and supplies were all purchased from Sigma-Aldrich Co. (St. Louis, MO, USA) and VWR International (Radnor, PA, USA). Triple-negative human breast (MDA-MB-231) cells were obtained from American Type Culture Collection (Rockville, MD, USA). Dulbecco’s modified Eagle’s medium (DMEM), fetal bovine serum (FBS) and penicillin/streptomycin were obtained from Invitrogen (Carlsbad, CA, USA). All microarray equipment, reagents and materials were purchased from Affymetrix/ Thermo Fisher (Waltham, MA, USA).

Cell culture. MDA-MB-231 cells were cultured in 75 cm² flasks containing DMEM supplemented with 10% FBS and 1% 10,000 U/mL penicillin G sodium/10,000 μg/mL streptomycin sulfate. Cells were grown at 37˚C with 95% atmosphere/ 5% CO₂ and sub-cultured every three to five days. YM-155 was dissolved in DMSO [20 mg/mL], and dilutions were prepared in sterile HBSS, ensuring solvent concentration of DMSO or absolute ethanol at less than 0.5%.

Cell viability assay. Alamar Blue cell viability assay was used to determine cytotoxicity. Viable cells reduce resazurin to resorufin, a fluorescence product. Briefly, 96-well plates were seeded with MDA-MB-231 cells at a density of 5×10⁶ cells/mL. Cells were treated with or without YM-155 for 24 h at 37˚C, 5% CO₂. Alamar Blue (0.1 mg/mL in HBSS) was added at 15% v/v to each well and the plates were incubated for 6-8 h. Quantitative analysis of dye conversion was measured using a Synergy™ HTX Multi-Mode microplate reader (BioTek, Winooski, VT, USA), at 550 nm/580 nm (excitation/emission). The data were expressed as viability: percentage of untreated controls.

Survivin protein expression. Human survivin ELISA: After treatment, cells were washed, lysed and evaluated for intracellular survivin protein expression using a survivin human SimpleStep ELISA^® Kit #ab183361 (Abcam, Cambridge, MA, USA) carried out in accordance with the manufacturer’s protocols. The data was quantified at 450 nm using a Synergy HTX multi-mode reader from Bio-Tek (Winooski, VT, USA).

Microarray WT 2.1 human datasets. After treatment, cells were washed three times in ice cold HBSS, rapidly frozen and stored at –80˚C. Total RNA was isolated and purified using the TRIzol/chloroform method, quality was assessed and concentration was equalized to 82 ng/μL in nuclease free water. Whole transcriptome analysis was conducted according to the GeneChipTM WT PLUS Reagent Manual for Whole Transcript (WT) Expression Arrays. Briefly, RNA was synthesized to first strand cDNA, second-strand cDNA and followed by transcription to cRNA. cRNA was purified and assessed for yield, prior to 2nd cycle single stranded cDNA synthesis, hydrolysis of RNA and purification of 2nd cycle single stranded cDNA. cDNA was then quantified for yield and equalized to 176 ng/mL. Subsequently, cDNA was fragmented, labeled and hybridized on to the arrays prior to being subject to fluidics and imaging using the Gene Atlas (Affymetrix, ThermoFisher Scientific). The array data quality control and initial processing from CEL to CHP files were conducted using expression console, prior to data evaluation using the Affymetrix transcriptome analysis console. Samples were run in triplicates (n=3). Supportive analysis was accomplished using geneontology.org (21). The data discussed in this publication have been deposited in NCBI’s Gene Expression Omnibus and are accessible through GEO Series accession number GSE103089 located at https://www.ncbi.nlm. nih.gov/geo/query/acc.cgi?acc=GSE103089.

Data analysis. Statistical analysis was performed using Graph Pad Prism (version 3.0; Graph Pad Software Inc. San Diego, CA, USA) with significance of difference between the groups assessed using a one-way ANOVA and then followed by Tukey post hoc means comparison test, or a Student’s t-test. LC₅₀ values were determined by regression analysis using Origin Software (Origin Lab, Northampton, MA, USA).

Results

The potent cytotoxic effects of YM-155 on MDA-MB-231 cells at 24 h were originally identified from a high-throughput (HTP) toxicity screening routinely conducted at our facility on thousands of compounds. HTP screenings involving tumor cell lines are always run in tandem with both positive and negative controls which are comprised of a large panel of reference chemotherapy drugs. For the sake of brevity, we provide data on a select few chemotherapy drugs vs. YM-155 (Figure 1), where YM-155 was more potent than doxorubicin, irinotecan, camptothecin and staurosporine in establishing lethality in MDA-MB-231 cells. Many of these screenings are conducted in a pilot manner to serve as a rationale for in depth investigation as in the case of YM-155.

Figure 1. Lead Hit Identification: A pilot screening of over 1,000 natural products, chemicals and drugs were evaluated for lethality in MDA-MB-231 cells at a 24-h time point. The data represent effects of sample reference chemotherapy drugs used in the screen, reflected by LC₅₀ calculated by dose response data acquired on a minimum of 6 concentrations, n=4. The most lethal compound at the lowest dose over 24 h was YM-155, a survivin inhibitor. These findings are in alignment with the literature.

In the next series of studies, we evaluated the loss of cell viability tantamount to the loss of survivin protein content. The LC₅₀ of YM-155 on MDA-MBA-231 cells was 14.6 ng/mL (32 nM) (Figure 2). Somatic survivin was first calculated using a standard curve, where the average survivin cell concentration was found to be 1177 pg/mL equating to 117.7 pg/50,000 cells (Figure 3A). Subsequently, YM-155 was capable of reducing survivin protein expression with an IC₅₀=2.1 ng/mL (Figure 3B). Given the slight drop in survivin, which preceded the loss in cell viability, the theoretical question arises as to if the drop in survivin could be a consequence of cell death, rather than a precipitating factor. Often time’s events occur that align with cell death, that may not itself be a factor to biological provocation. To further investigate this aspect, the study was repeated in another TNBC cell line: the MDA-MB-468 cell line. The data acquired on the MDA-MB-468 cell line was even less remarkable, showing the IC₅₀ for survivin at 16.9 ng/mL corresponding to loss of viability with an LC₅₀ at 34.35 ng/mL (data not shown). The loss of survivin in both cases was only observed at concentrations slightly preceding cell death. To further investigate if meager changes in the survivin protein by YM-155 were reflected at the genomic level, samples were analyzed for changes to the whole transcriptome (WT) by genomics microarray methodology.

Figure 2. YM-155 toxicity in MDA-MB-231 cells at 24 h. The data represent cell viability as % of untreated controls and are expressed as the Mean± S.E.M, n=4. Significant of differences from the control were evaluated using a one-way ANOVA, followed by a Tukey-Post Hoc Test, *p<0.05. The LC₅₀ of 14.6 ng/mL is equivalent to 32 nM. n= Concentration used for WT microarray evaluation.

Figure 3. (A) ELISA standard curve for human survivin. l denotes the average expression of survivin in untreated MDA-MB-231 cells=1,177 pg/mL equating to 117.7 pg/50,000 cells prepped in 100ul of lysate buffer. (B) Intracellular concentration of survivin in YM-155 treated MDA-MB-231 cells at 24 h. The data represents survivin concentration as % of untreated controls and are expressed as the Mean±S.E.M, n=3. Significance of differences from the control were evaluated using a oneway ANOVA, followed by a Tukey-Post Hoc Test, *p<0.05.

RNA was collected from cells treated with 14.6 ng/mL of YM-155 at 20 hours, where the data show a meager reduction in survivin (-2.7 FC, p<0.001). The change in survivin was largely overshadowed by changes in 2,592 transcripts, many of which had greater FC and significance (Figure 4). Given that cell death alone may constitute an interfering variable to WT data, a second WT data set was obtained at an earlier time point (8 hours) where death had not occurred yet, and concentrations were adjusted for sub-lethal/high concentration (83 ng/mL) (Figure 5). This concentration was high enough to trigger what would be a complete loss of cell viability at 24 hours, suitable for examining events precipitating at the cusp of cell death.

Figure 4. Whole transcriptome differential shift by YM-155 (14.6 ng/mL) in MDA-MB-231 cells at 20 h, criteria set at: FC +/-2, p-Value<0.05. The data is presented as Bi-weighted Ave Signal (Log2) of raw intensity values, with expression (low) to (high) and distance from the central diagonal line being FC down (red) or up (green). Amidst the differentially expressed genes is n the survivin transcript, -2.67 FC, ANOVA p-Value=0.0000907, n=3.

Figure 5. Sub-lethal YM-155 concentration setup for WT microarray. Optimum concentration was 83 ng/mL which at 8 h showed little effects on viability, but was high enough to trigger complete cell death over the next 16 h. The data represents cell viability as % untreated controls and are expressed as the Mean±S.E.M, n=4. Significance of differences from the control values were evaluated using a one-way ANOVA, followed by a Tukey-Post Hoc Test, *p<0.05. n= Concentration used for WT microarray evaluation.

Transcriptome differentially expressed transcripts are presented for both 8 and 20-h time points in the form of a volcano plot showing FC vs. significance (p-value), from triplicates n=3 (Figure 6A and B). Figure 6A and B shows a high degree of overlap for specific transcripts at both time points, these transcripts are also being presented in Table I. Table I shows only the transcripts common to both time points, where the entire data series for control vs. 8 h, and control vs. 20 h are available at: NCBI’s Gene Expression Omnibus and are accessible for public analysis through GEO Series accession number GSE103089 located at https:// www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE103089 Again, these findings clearly show a plethora of changes occurring tantamount to the loss of the survivin transcript, suggesting this is unequivocally not a single target drug.

Figure 6. (A, B) Time-dependent effects on transcripts of YM-155-treated MDA-MB-231 cells: (A) for 8 h (83 ng/mL) (B) for 20 h (14.6 ng/mL). The data are plotted as a volcano, showing both FC (X-Axis) <2x - (negative)/red and - >2x (positive)/green and significance on the Y-Axis, n=3. Highlighted changes for specific genes describe those altered at both time points. These are likely to be most influential in YM-155 mediated toxicological effects.

Table I. The data represent Gene Symbol, description, cluster ID, Fold-Change, ANOVA p-value and FDR p-values.

The spot intensity bi-weighted average signal (log2) for survivin quantified on all 9 oligonucleotide arrays is presented in Figure 7. Figure 7 shows reduction of BIRC5 as a minor shift relative to the vast predominating effects of YM-155 which by 20 h had significant impact on the following processes: DNA replication (Figure 8), DNA damage (Figure 9), cell cycle (Figure 10) and apoptosis (Figure 11). Down-regulated transcripts are denoted by solid red, up-regulated transcripts by solid green, and ///// = no statistical differences.

Figure 7. Survivin mRNA transcript change in YM-155 treated cells 8 h (83 ng/mL) and 20 h (14.6 ng/mL). The data represent raw data triplicate microarray spot values by intensity presented as bi-weighted average Signal (log2) for the BIRC5/baculoviral IAP repeat containing 5 gene.

Figure 8. DNA REPLICATION: YM-155 affected pathway elements at 20 h (14.6 ng/mL) in MDA-MB-231 cells. Down-regulated transcripts are denoted by solid red, up-regulated transcripts by solid green, and ///// no statistical differences.

Figure 9. DNA DAMAGE: YM-155 affected pathway elements at 20 h (14.6 ng/mL) in MDA-MB-231 cells. Down-regulated transcripts are denoted by solid red, up-regulated transcripts by solid green, and ///// no statistical differences.

Figure 10. CELL CYCLE: YM-155 affected pathway elements at 20 h (14.6 ng/mL) in MDA-MB-231 cells. Down-regulated transcripts are denoted by solid red, up-regulated transcripts by solid green, and ///// no statistical differences.

Figure 11. APOPTOSIS: YM-155 affected pathway elements at 20 h (14.6 ng/mL). Down-regulated transcripts are denoted by solid red, up-regulated transcripts by solid green, and ///// no statistical differences.

These findings show major changes by YM-155 to invoke major loss to the efficacy of normal DNA damage response, which is likely to be a formidable and primary mechanism of action for this drug.

Discussion

Survivin is an inhibitor-of-apoptosis protein with key roles in driving uncontrolled cell division in aggressive malignancies. Small-molecule survivin inhibitors are a sought after chemotherapy drug classification, to which sepantronium bromide (YM-155) is believe to be best in in class. (22-24) YM-155 has been reported to overcome drug resistance (e.g. statins (16,25-27) docetaxel (28)), improve efficacy of diverse therapies including hypericin-mediated photodynamic therapies (29) and on its own merit, induces apoptosis, reduces colony formation, attenuates invasion and growth of xenograft tumors in nude mice (30-33). While reduction of survivin by YM-155 is believed to be a foremost mechanism of action, other biochemical targets for this drug are continually being reported in the literature. Just for example, YM-155 has been reported to alter transcription of FOS, JUN and c-MYC (34), pro-apoptotic Bak (35), p53 (22), death receptor 5, other IAP family member proteins (BIRC2, MCL-1), PI3K-Akt (31) and autophagy regulating components such as Beclin1 (36-38). In this study, we provide supportive information as to the diverse transcriptome targets of YM-155 in TNBC cells while maintaining a benchmark differential comparative in survivin protein and mRNA changes.

The data in this study show only a small reduction of survivin within 8 h of treatment, despite enormous global changes to the transcriptome, some of which are exacerbated by 20 hours coinciding with cell death. Table I shows only the changes evident in BOTH groups, where as individual datasets for control and treatment (8 h) and (20 h) have been deposited with NCBI’s Gene Expression Omnibus which are accessible through GEO Series accession number GSE103089 located at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE103089.

Up-regulation of tumor suppressors. YM-155 elicited enormous differential transcript changes some consistently observed at both 8 and 20-hour time points to which many of these correspond to under-researched transcripts of unknown function. Some of these include the YM-155 mediated up-regulation of tumor suppressor: FBJ murine osteosarcoma viral oncogene homolog B (FOSB) [+12.42, p<0.001, 8 h]. While little is known about FOSB, it is reportedly under-expressed in malignant tumors (39,40) with lower levels predicting aggressive metastasis, cell proliferation, clonal formation and overall poor patient survival rates (41). In contrast, the up-regulation of FOSB would afford greater chemo-sensitivity to tumoricidal peptides or anthracyclines (42) and favorable recurrence-free survival in cases of breast cancer (43). More research will be required to investigate not only a role for FOSB in diverse cancers, but also how YM-155 can influence over this gene. Another example of a relatively under-studied transcript up-regulated by YM-155, is salt inducible kinase 1 (SIK1) which is evident at both time points [+5.04, p<0.001, 8 h/ +8.06, p<0.001, 20 h]. Like FOSB, SIK1 has been reported as a tumor suppressor capable of halting tumor growth, metastasis and rendering greater overall survival rates (44,45). Likewise, numerous studies are now reporting that a reduction of SIK1 is associated with drug/adiation resistance, aggressive tumor pathologies and greater likelihood of epithelial–mesenchymal transition (46-48). Again, more work is needed to investigate not only a role for SIK1 in diverse cancers, but the YM-155 influence over this gene. Yet, another example of a YM-155 target gene to which little is known, is that of up-regulated expression of tumor necrosis factor, alpha-induced protein 3 (TNFAIP3) [+18.08, p<0.001]. What we do know from previous research is that low levels of TNFAIP3 are related to germline mutations associated with a plethora of inflammatory diseases (49) such as arthritis, myasthenia gravis (50) autoimmune hepatitis (51) autoimmune lymphoproliferative syndrome (52) and T-cell acute lymphoblastic leukemia (53,54). TNFAIP3 functions an endogenous anti-inflammatory regulator with capacity to down-regulate nuclear factor kappaB signaling, (52) attenuate suppressor of cytokine signaling 3 (55) and JNK phosphorylation (56), with its overexpression triggering tumor apoptosis, and reduction in cytokine release as well as metastasis/migration/ invasion (57-60).

Future research on many of these transcripts (Table I) is required to unveil the most influential aspect of this drug. The boundaries of this discussion will now focus on the known major pathways affected by YM-155 in MDA-MB-231 cells at the 20 hr time point. The data shows significant impact primarily on reduction in genomic stability, with broad scale changes in pathway elements associated with DNA replication, DNA damage response pathways (DDR) and cell cycle (Figure 9-11) rather than a loss of survivin and its role in apoptosis (Figure 11).

DNA damage pathway. YM-155 caused numerous down-regulated processes at 20 h which are otherwise required to overcome damaged DNA at the stalled replication fork through two central kinase controlled signaling pathways 1) ATR (ATM- and Rad3-Related) and 2) ATM (ataxia-telangiectasia mutated) (Figure 9). Loss of function in both ATR and ATM would render faulty repair of replication-associated DNA damage and DNA double-strand breaks, to which YM-155 mediated a significant reduction key player, such as Fanconi complementation group D2 (FANCD2) [–7.16, p<0.001]. This particular gene is linked to the control of hypersensitivity to the damaging effects of crosslinking agents of DNA. A genetic inherited recessive mutation of FANCD2 in the germline creates major chromosomal instability where FANCD2^–/– or a single mutation in BRCA2 [–5.98, p<0.001] are associated with replicative DNA errors and greater risk of diverse cancers including ovarian, head and neck and breast (61). Once cancers are established, anti-cancer drugs that impair the damaged DNA response pathways such as DNA interstrand crosslinking (ICL) agents are employed like cisplatin, mitomycin C or related chemotherapy drugs (62). The results of this study suggest this could on of the primary mechanisms for YM-155 mediated lethal effects on breast cancer cells.

Moreover, according to Figure 9 and the GEO 20 h full dataset: the data show a YM-155 mediated reduction in not only FANCD2 expression – but also many of the Fanconi anemia E3 monoubiquitin ligase core complex components (FANCA/B/C/E/F/G/L/M): group A (FANCA [–4.03, p<0.001], group B (FANCB [–3.34, p<0.001]), group L ANCL [–4.21, p<0.001]), group E (FANCE [–2.02, p<0.001]), group F (FANCF [–2.17, p<0.001], group G (FANCG [–2.28, p<0.001]) group M (FANCM [–3.62, p<0.001]) all of which are needed to activate FANCD2 through ubiquitination. A loss of ubiquitinated FANCD2 renders inadequate formation of the FANCD2/FANCI complex [–8.56, p<0.001] which is needed to initiate DNA repair by driving formation of the multicomplex BRCA1 [–3.14, p<0.001], BRAC2 and RAD51 [–3.74, p<0.001], which were themselves also affected. Just a single target impact on one of these systems alone (e.g. FANCD2 monoubiquitination) can augment existing chemotherapies of the ICL class (61,63). YM-155 appears to be a multi-target drug with major effects on DNA repair.

Even further, in regards to DNA damage response, YM-155 also caused reduction in downstream elements of the ATM/FANCD2 pathway. An example of this is the reduced transcript for PRKDC [–3.3, p<0.001], which would fail to activate CHEK2 to phosphorylate and activate BRAC1. Failure in this particular step would prevent complex formation of ubiquitinated FANCD2 to BRCA1 and RAD51 and halt the DDR. Not only this, but YM-155 also adversely affects the partner and localizer of BRCA2 (PALB2 [–2.33, p<0.001]), which is required for protein linking BRCA1, BRCA2 and RAD51 at the site of a DNA double-strand break. This is pivotal because a combined reduction of BRCA1 interacting protein C-terminal helicase 1, BRIP1 [–6.37, p=6.43E-07] (FANCJ) at both time points would prevent its role as a major helicase needed to interact with BRCA1 to oversee homologous recombination DNA repair and cell response to replication stresses (64). Although many more targets are centered around YM-155 mediated impairment to the DNA damage response pathway, as a last example, we see reduction in exonuclease 1 (EXO1 [–4.57, p<0.001]) which is required for 5’ resection in DNA mismatch repair processes involving homologous recombination repair at stalled replication forks (65-67). Lastly, YM-155 simultaneously down-regulated up/stream and downstream targets of EXO1 such as ATM, ATR and cyclin-dependent kinases 1[–4.09, p<0.001] and 2 findings which all corroborate major impact on DNA damage response. A significant loss of these processes would augment effects of chemotherapy, radiotherapy, and poly(ADP-ribose) polymerase inhibitors (68). These findings suggest a very potent role of YM-155 to destabilize integrity of the genome through direct genetic repression of Fanconi anemia pathways transcript components – all of which would lead to inability to repair DNA (69).

DNA replication. YM-155 also influenced down-regulation of transcripts associated with the assembly of the pre-replicative complex/CDC6 [–2.82, p<0.001], GMNN [–2.18, p<0.001] and several mini-chromosome maintenance helicases (MCM) transcripts. Cell cycle protein CDC6 is involved with recruiting MCMs 2-7 onto the origin recognition complex (ORC), then licensed for replication (70,71). YM-155 caused losses in processes associated with the activation of the ORC in particular by down-regulating CDC7 [–3.95, p<0.001] needed to phosphorylate MCM proteins during the S phase for activation of the replication initiation complex assembly (72,73). Losses of POLA2 [–2.74, p<0.001] would render the reduction in DNA synthesis, which could be beneficial to augment effects of chemotherapies such as gemcitabine in resistant cell lines (74). YM-155 also caused simultaneous reduction in the polymerase complex (POLA/primase) PRIM1 [–2.17, p<0.001] and POLE2 [–2.68, p=0.004308] which could lead to faulty telomere processing (75). In addition, YM-155 triggered negative changes in the RNA primer-DNA primer with losses in replication factor C (RFC) including both the large subunit [replication factor C, subunit 1 (RFC1)] [–2.14, p<0.001] and four small subunits [replication factor C, subunits 2-5 (RFC25)], which could lead to impaired activation of DNA polymerases [6] through inadequate loading of PCNA onto chromatin and DNA polymerases to PCNA (76-78), Several of these changes overlap onto the YM-155 effects on cell cycle (Figure 10).

Cell cycle. In response to DNA damage, cells are arrested through replication stress by down-regulating the activity of cyclin-dependent kinases (CDKs) and cell cycle kinases, including Polo-like kinase 1 (PLK1) [–2.41, p<0.001], Bora [–4.56, p<0.001] and Aurora kinases (79). In this study, YM-155 mediated a loss of many of these processes including CCNB1 [–2.5, p<0.001] and CCNB2 [–2.77, p<0.001] (which encodes G2/mitotic-specific cyclin-B 1 and 2 proteins) cyclin B/CDK1, BORA, AURKB, CENPA, CCNE2 and its target PLK1 [–2.41, p<0.001], all of which are essential for mitotic recovery after DNA damage (80-85). Many of these cycle checkpoint transcripts are reportedly overexpressed in aggressive chemo resistant tumors (86-91) leading to unbridled proliferation (92), poor prognosis (93) to which small-molecule inhibitors that target these kinases are currently being tested as anticancer drugs (94-96). Kinase inhibition would indirectly prevent Plk1 activation, initiate G2 arrest (97), trigger a simultaneous rise in tumor suppressors- p53-inducible p21 tumor, Wee1, Myt1, p21 (98), as well as proteasomal Bcl-2 degradation (99) all contributing to cytostatic mediated apoptosis (100-107). YM-155 appears to down-regulate this mitotic cascade through attenuating transcript levels of related kinases.

The data from this work also shows that YM-155 mediates a loss in (budding uninhibited by benzimidazoles 1) BUB1 [–5.95, p<0.001]/BUB1B which could block not only checkpoint activity leading to mitosis but also spindle assembly, kinetochore–microtubule attachment and chromosome segregation (88,108,109). Other related transcripts down-regulated by YM-155 include SKP2 [–2.23, p<0.001] which encodes for S-phase kinase-associated protein 2 which is part of SCF complex (SKp1-cullin-FBox)/ Cks1 inversely associated with concentration of p27Kip1 and p57Kip2 (CDK inhibitors) which initiate cell cycle arrest (110,111).

Clinical Application

While the data in this study support existing literature describing the lethal effects of YM-155 on tumor cells in vitro, its clinical application remains uncertain. The National Institutes of Health website, ClinicalTrials.gov, lists 11 clinical trials using YM-155. Most of these are phase I and II trials assessing tolerability and dosing with few powered to study clinical efficacy against cancer. Phase II trials evaluating adjunctive use with standard treatments such as rituximab (relapsed aggressive B-cell non-Hodgkin lymphoma), carboplatin and paclitaxel (lung cancer) or docetaxel (HER2 negative breast cancer) show adequate tolerability with moderate to no clinical improvements with side effects including alopecia, fatigue, nausea, decreased hemoglobin/anemia, headache and neutropenia (112-117). It appears unlikely that YM-155 will be useful as an add-on medication in advanced late-stage metastatic disease. Perhaps a clearer understanding the actions of this drug, higher levels of tolerated dose and understanding its biological stability will allow a rational deployment in which it has efficacy.

In conclusion, the data from this work provide a framework circumscribing the entire transcriptome differential response to YM-155 in TNBC cells. This data can serve as a reference point for future work on defining a more finite and concrete mechanism for this drug. Clearly, YM-155 exerts dynamic global effects on the transcriptome, affecting hundreds if not thousands of processes independent of survivin and inclusive to major negative impact on DNA replication, cell cycle and DNA damage would exert negative controlling influence over tumor growth and augment effects of ICL chemotherapy agents (63,118). The findings in this study can be further explored using the public dataset, and functional biological algorithms in NCBI’s Gene Expression Omnibus and are accessible through GEO Series accession number GSE103089 located https:// www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE103089.

Conflicts of Interest

The Authors wish to confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial support for this work that could have influenced its outcome.

Acknowledgements

This project was supported by the National Institutes of Health, National Institute on Minority Health and Health Disparities, RCMI grant (8G12MD007582-28.) and COE grant (P20 MD006738). We also wish to thank Dr. Shawn Lee Thomas and students: Davonta Powe and Anthony A Jumelles from the Dept. of Biology / Florida A&M University for their support and assistance.