Topoisomerase IIα in chromosome instability and personalized cancer therapy.

Genome instability is a hallmark of cancer cells. Chromosome instability (CIN), which is often mutually exclusive from hypermutation genotypes, represents a distinct subtype of genome instability. Hypermutations in cancer cells are due to defects in DNA repair genes, but the cause of CIN is still elusive. However, because of the extensive chromosomal abnormalities associated with CIN, its cause is likely a defect in a network of genes that regulate mitotic checkpoints and chromosomal organization and segregation. Emerging evidence has shown that the chromosomal decatenation checkpoint, which is critical for chromatin untangling and packing during genetic material duplication, is defective in cancer cells with CIN. The decatenation checkpoint is known to be regulated by a family of enzymes called topoisomerases. Among them, the gene encoding topoisomerase IIα (TOP2A) is commonly altered at both gene copy number and gene expression level in cancer cells. Thus, abnormal alterations of TOP2A, its interacting proteins, and its modifications may have a critical role in CIN in human cancers. Clinically, a large arsenal of topoisomerase inhibitors has been used to suppress DNA replication in cancer. However, they often lead to the secondary development of leukemia because of their effect on the chromosomal decatenation checkpoint. Therefore, topoisomerase drugs must be used judiciously and administered on an individual basis. In this review, we highlight the biological function of TOP2A in chromosome segregation and the mechanisms that regulate this enzyme's expression and activity. We also review the roles of TOP2A and related proteins in human cancers, and raise a perspective for how to target TOP2A in personalized cancer therapy.

INTRODUCTION

Chromosome instability (CIN) is one of the hallmarks of cancer cells[1] and is often mutually exclusive from hypermutation genotypes. Hypermutations in cancer cells are due to defects in DNA mismatch repair genes, including MLH1 and MSH2;[2] however, the cause of CIN is still elusive. Most solid tumor cells with CIN are aneuploidy, indicating that abnormal mitosis is involved in the CIN phenotype. The process of mitosis is precisely regulated. The sequence of events is divided into six stages (prophase, prometaphase, metaphase, anaphase, and telophase) corresponding to the completion of one set of activities and the start of the next. During mitosis, the pairs of chromatids condense and attach to spindles that pull the sister chromatids to opposite sides of the cell, each cell receives chromosomes that are alike in composition and equal in number to the chromosomes of the parent cell. The decatenation and mitotic checkpoints govern sister chromosome segregation. Topoisomerase IIα (TOP2A) is a key player in the decatenation checkpoint; when the decatenation checkpoint is defective, chromosomal mis-segregation is observed. CIN can provide these evolving cell subclones with a mechanism that causes unremitting genomic and mutational plasticity. Furthermore, the decatenation checkpoint deficiency can cause additional chromosome imbalances in cancer cells, increasing tumour aggressiveness (Figure 1).

Figure 1

Cancer biology of CIN

Normally, a 4N cell in G2 enters mitosis, aligns its chromosomes in the metaphase plate, and equally distributes the DNA over two nuclei and subsequently, two daughter cells.[30] The decatenation checkpoint controls sister chromosome segregation and delays entry into mitosis until the knotted chromosomes have been decatenated by TOP2A. When the decatenation checkpoint is defective, chromosomal mis-segregation is observed. This causes aneuploidy and CIN. CIN can provide these evolving cell subclones with a mechanism that fosters unremitting genomic and mutational plasticity; this leads to increased cell survival, increased cell proliferation, and carcinogenesis. Furthermore, the decatenation checkpoint deficiency can result in additional chromosome imbalances of cancer cells, increasing tumor malignancy. CIN can slow the proliferation rate and cause cell death in cells with chromosomal alternations but no unremitting genomic and mutational plasticity.

In this review, we discuss the biological function of TOP2A in chromosome segregation and CIN. We also assess the mechanisms that may regulate this enzyme's expression and activity and shed light on the roles of TOP2A in human cancer progression and personalized cancer therapy.

CHROMOSOME INSTABILITY AND CANCER

CIN involves the unequal distribution of DNA to daughter cells upon mitosis, resulting in the loss or gain of chromosome during cell division, and ultimately, aneuploidy. Aneuploidy can be driven by genetic alterations that promote inaccurate chromosome segregation, which often causes whole chromosomes to be lost or gained. This condition is termed whole chromosome instability. In addition to numerical chromosomal abnormalities, cancer cells can show changes in chromosome structure, such as deletions of chromosome arms and amplifications of large chromosome regions. This condition is termed structural or segmental chromosomal instability.[3]

Recently, CIN was recognized as a distinct feature of most aggressive cancer types (Figure 2). In normal organisms, the integrity of the chromosome number and structure are essential to maintaining survival. Aneuploidy, in the form of large or local chromosomal changes, is more evident and unique in human cancer. Aneuploidy is distinct from polyploidy; in polyploidy, cells contain more than two complete sets of chromosomes but always have an exact multiple of the haploid number, so the chromosomes remain balanced. Cancer cells not only have numerous gene mutations but are also characterized by aneuploidy and CIN.[2]

Figure 2

CIN-related gene aberrations in solid tumors

CIN was recognized as a distinct feature of most solid tumors that have numerous CIN-related gene aberrations. The inner circle shows various cancer types. The outer circle shows related genes that cause CIN.

Whether CIN is a driver or passenger of cancer is a subject of debate in cancer research. Most researchers believe that CIN is an early event in tumor formation that causes the deletion or inactivation of tumor suppressor genes through defects in genome maintenance and destabilization of the nucleotide sequence;[1,4,5] other researchers argue that CIN is a side effect of neoplastic growth and cell division, during which cancer cells frequently lose and gain chromosomes.[6]

To better define the role of CIN in cancer, it is important to understand the molecular basis of tumorigenesis. Genome alterations in cell subclones confer advantages on outgrowth and lead to the subclones’ dominance in the local tissue environment. CIN is a cell-autonomous feature that provides evolving cell subclones with a mechanism that causes unremitting genomic and mutational plasticity, leading to increased cell survival, cell proliferation, and tumorigenesis.[7] In essence, CIN drives tumor progression by accelerating the gain of oncogenic loci and the loss of tumor suppressor loci. Therefore, other subclones that have chromosomal alternations but do not have this kind of gain and loss will not acquire tumor-promoting features. Finally, multistep tumor progression can proceed because of the succession of subclone expansions. Comparative genomic hybridization, which can document the gains and losses of gene copy numbers across the cell genome, provides clear evidence of the loss of control of chromosome integrity in tumor progression.[8] The development of next-generation DNA-sequencing technologies has led to the detection of widespread destabilization of gene copy numbers and nucleotide sequences.[9,10] Defects in genome maintenance systems cannot detect and resolve these abnormal genome alterations[1]. Therefore, favorable genotypes accumulate in subclones or premalignant cells. Accordingly, CIN is likely the primary cause and driving force of tumorigenesis and progression.[11]

CHROMOSOME SEGREGATION AND CIN

The chromosomal changes that cause CIN commonly occur during tumor cell division.[2] Over 100 genes are involved in sister chromosome segregation in humans, and defects in some of the genes that control this segregation are found in cancer cells.[6,12] The mitotic checkpoint plays a prominent role in chromosome segregation. This checkpoint is a control mechanism that ensures high-fidelity chromosome segregation. One of the key steps is regulated by the anaphase-promoting complex/cyclosome (APC/C). APC/C is a multi-subunit E3 ubiquitin ligase that controls mitotic progression and sister chromosome segregation. APC/C activation requires cell-division-cycle 20 (CDC20) as a co-activator to modulate the ubiquitination and degradation of mitotic substrates, including securin and cyclin B1.[13] Degradation of securin releases the sequestering partner, separase. Active separase can cleave the cohesin links that hold the sister chromatins together. Degradation of cyclin B1 leads to the inactivation of cyclin-dependent kinase 1 and initiates mitotic exit. Numerous studies have determined the effect of the reduced expression of mitotic checkpoint genes, including Bub1, Bub1b, Bub3, Mad1, Mad2, Cenp-E, Cmt2, Mps1, ZW10, ROD, and Zwilch, and found that the haploinsufficiency of these genes resulted in chromosome mis-segregation (Table 1 and references within). Several key proteins, including TOP2A, cooperate with APC/Ccdc20 and regulate sister chromosome segregation (Figure 3).

Table 1

Proteins involved in chromosome segregation

Protein	Description	Function	Ref.
Bub1	Budding uninhibited by benzimidazole 1	Inhibit CDC20 by phosphorylation	14,15
Bub1b	Bub1p	Encoding Bub receptor 1 protein	16
Bub3	Budding uninhibited by benzimidazole 3	Localize Bub1 and BubR1 to kinetochores	17
Mad1	Mitotic arrest deficient 1	Recruit Mad2 to unattached kinetochores	18
Mad2	Mitotic arrest deficient 2	Bind to CDC20 and inhibit APC/C activity	19,20
Cenp-E	Centromere protein E	Activate Bubr1 at the unattached kinetochore	21
CMT2	Charcot-Marie-Tooth gene 2	Inhibit mitotic checkpoint signaling by antagonizing MAD2	22,23
MPS1	Monopolar spindle 1	Phosphorylate Bub1	24,25
ZW10	Zeste white 10 protein	Recruit the MAD1–MAD2 heterodimer to unattached kinetochores	26,27
ROD	Rough deal protein	Recruit the MAD1–MAD2 heterodimer to unattached kinetochores	26,28
Zwilch	N/A	Recruit the MAD1/2 heterodimer to unattached kinetochores	26,29
Securin	N/A	Activate separase	30
Separase	N/A	Cleave the cohesion links	31,32
CDC20	Cell-division cycle protein 20	Cofactor of APC/C	33,34
Cdh1	Cadherin-1	Cofactor of APC/C	35,36
TOP2A	Topoisomerase IIalpha	Decatenation checkpoint, separate knotted and intertwined DNA molecules, decatenate intertwined daughter DNA duplexes	37-39
Chfr	Checkpoint with Forkhead and RING finger domains	E3 ubiquitin ligase, inhibit cyclin B nuclear import, regulate Mad2 and BubR1 functions	40
Nup98	Nucleoporin 98	Prevent cohesion degradation	35
Rae1	Bub3-related protein RNA export factor 1	Prevent cohesion degradation	35
RanBP2	Ran binding protein 2	Promote disentanglement of sister chromatids	37
Plk1	Polo-like kinase I	Phosphorylate and remove cohesion complex proteins	41,42
Sgo1	Shugoshin 1, inner centromere protein	Counteract phosphorylation of cohesion; may directly inhibit separase	43
Incenp	Inner centromere protein	Localize along chromosome arms in anaphase	44,45
Cdc5	Polo-like kinase	Phosphorylate and remove meiotic cohesin	46
Survivin	Inhibitor of apoptosis protein	Unclear; may play multiple roles in regulating apoptosis and cell division	47
MCAK	Mitotic centromere-associated kinesin	Coordinate onset of sister centromere separation	48
Orc6	Origin recognition complex 6	Coordinate chromosome replication and segregation	49
MIIP	Migration and invasion inhibitor protein	Interact with CDC20 and inhibit APC/C activity	50

Figure 3

Regulation of sister chromatid separation

In pro-metaphase, various mitotic checkpoint proteins, including Mps1, Rae1, Bub1, Bub3, BubR1, Mad1, and Mad2, bind kinetochores that lack attachment or tension. Unattached chromosomes apparently generate a signal that delays progress to anaphase until all sister chromatids are attached to the spindle apparatus. This signal is transduced by the spindle-checkpoint complex that includes CENPE and the Mad/Bub proteins, resulting in inhibition of APC/CCdc20. After attachment of the last kinetochore to the mitotic spindle, the “wait anaphase” signal is extinguished. This allows APC/CCdc20 to become active, resulting in the ubiquitin-dependent degradation of cohesion. The separase is then activated, and the protease catalyzes the cleavage of cohesion complexes that contain Rec8, which bridges the aligned sister chromatids. The kinase Plk1 marks Rec8 at chromatid arms by phosphorylation so that Rec8 will be degraded by separase, while a complex of Sgo1 is recruited by Bub1 to centromeres. Sgo1 can bind and protect Rec8 from phosphorylation by Plk1. Sister chromatid separation is also dependent on TOP2A, which targets inner centromeres and decatenates centromeric DNA. The newly separated chromatids can then migrate poleward along the spindle axis during anaphase. Once the chromosomes become bi-oriented, Sgo1 is dephosphorylated at Thr-346 and Sgo1 no longer binds to cohesion. NMSP715, NMS-P937, BI6727, BI2536, UA62784, etoposide, and teniposide target the mitotic checkpoint for cancer therapy.

ROLES OF TOP2A IN CIN

TOP2 is a group of highly conserved enzymes that catalyze the ATP-dependent transport of one intact DNA double helix through another.[51] As a result, the intertwined parental strands of a replicating DNA ring can come apart, interlocked double-stranded DNA rings can become unlinked, and knots can be introduced or removed from DNA rings.[52] Previous studies have revealed the important role of TOP2 activity in chromosome segregation through the use of inhibitors. Drugs that interfere with TOP2 have been reported to induce polyploidy and endoreduplication to different degrees, providing indirect evidence that this enzyme is required for the separation of sister chromatids.[53-58]

Human somatic cells replicate and segregate their genomes with remarkable precision. In the cell cycle, sister chromatids become entangled after DNA replication.[59] Normally, the decatenation checkpoint monitors the chromosome catenation status. Cells are arrested at the G2 phase, and the onset of mitosis is delayed if sister chromatids are not fully separated.[60-62] Because complete chromatid decatenation is required for accurate chromatid segregation, it has been suggested that the attenuation of the decatenation checkpoint function contributes to the acquisition of CIN in cancer cells.[61] TOP2A possess three distinct subunit dimerization interfaces (Figure 4A) and is maximally expressed in the G2 and M phases of the cell cycle.[63-66] It is specific for chromosome untangling and is essential for sister chromosome segregation before anaphase.[67-69] One biochemical study performed in Xenopus egg extracts (XEEs) showed that centromeres were catenated while sister chromatids underwent bipolar attachment and that TOP2A activity was required after the onset of anaphase.[69] As an alternative approach, TOP2 catalytic inhibitors have been used to demonstrate the role of TOP2A in sister chromosome segregation. TOP2 catalytic inhibitors determine the relative importance of the enzyme in promoting chromosome segregation at the metaphase-anaphase transition. Inactivation of TOP2 by merbarone also resulted in polyploidy in male mouse meiotic cells.[70]

Figure 4

Regulation of TOP2A

A) Domain arrangement of TOP2A. TOP2A possesses 3 distinct subunit dimerization interfaces, termed the N-gate, DNA-gate, and C-gate. Functional regions and post-modification sites are colored and labeled. GHKL, GHKL (gyrase, Hsp90, histidine kinase, MutL) family ATPase; TOPRIM, Mg2+ ion-binding Topoisomerase/Primase fold; WHD, winged helix domain; CTR, C-terminal domain. B) Transcriptional regulation of TOP2A expression. The expression of human TOP2A is controlled by its promoter region that contains two GC boxes and five CCAAT boxes. NF-Y recognizes and binds to the ICBs. This binding of NF-Y to the TOP2A promoter can be promoted by HMGB1/2 and inhibited by pRb. In the TOP2A promoter, GC1 and GC2 flank ICB1 and ICB5, respectively. Two specificity proteins, Sp1 and Sp3, regulate TOP2A transcription by binding to both GC1 and GC2. Sp1 is a transcriptional activator and can up-regulate TOP2A transcription, while Sp3 is a transcriptional repressor of TOP2A and a common modulator of Sp1-dependent transcriptional activation. C) Post-translational modifications of TOP2A. TOP2A is activated by phosphorylation and enhanced by HDAC1 and HDAC2, but it is inhibited by the E3 ubiquitin ligase activity of BRCA1. SUMO modification, which is catalyzed by RanBP2, leads TOP2A to accumulate at inner centromeres and is essential for proper sister chromosome separation in mitosis. P, phosphorylation; S, SUMOylation; T, TOP2A.

TOP2 catalytic inhibitors inhibit the ATPase activity of TOP2A and stabilize this enzyme in a closed-clamp form, rather than stabilizing the TOP2A DNA-cleavable complex, which is the mechanism of action of TOP2 poisons (e.g. etoposide and teniposide).[71] Therefore, in contrast to TOP2 poisons, TOP2 inhibitors do not induce extensive DNA breaks. Among the classes of catalytic TOP2 inhibitors, the bisdioxopiperazines (e.g., ICRF-154, ICRF-187, and ICRF-193) have been the most extensively studied.[72,73] Andoh et al reported that ICRF-193, a catalytic, noncleavable-complex-forming-type TOP2 inhibitor, led to an absence of chromosome segregation at mitosis, with further accumulation of polyploid cells.[74] In addition, treating human leukemia cells with ICRF-187 led to endoreduplication, which resulted in large and highly polyploid cells.[75] However, these TOP2 inhibitor studies did not reveal whether a single isoform was responsible, and these phenotypes may have been complicated by side effects of the inhibitors. Gene targeting in mice showed that segregation was dependent on the alpha subunit of TOP2, not the beta subunit of TOP.[76,77] When TOP2A's function was blocked after chromosome condensation, cells arrested at metaphase, chromosomes failed to separate, and anaphase bridges formed,[53,57,78,79] resulting in partial or complete chromosome gains or losses and polyploidy; this observation supports the theory that the enzyme is important in anaphase segregation.[80,81]

As a whole, these reports support the theory that the catenation state of intertwined sister chromosomes is monitored in G2 cells and that progression to mitosis is actively delayed when chromosomes are not sufficiently decatenated. The final step, decatenation of intertwined daughter molecules, can only be carried out by TOP2A.

TOP2A EXPRESSION REGULATION

TOP2A expression peaked in G2/M phase cells and decreased when cells completed mitosis. Cell cycle-dependent TOP2A expression is essential, and TOP2A depletion in mammalian culture cells causes severe defects in chromosome segregation during anaphase.[82] The expression level of human TOP2A is controlled by its promoter region. The TOP2A promoter does not contain a consensus TATA motif but contains two GC boxes and five CCAAT boxes that are located mostly in an inverted orientation (Figure 4B). The activity of the TOP2A promoter is regulated by various external stimuli, including the stages of the cell cycle, and by the TP53 tumor suppressor protein. Experimental studies using cell lines showed that TOP2A expression was negatively regulated by wild-type TP53 through its promoter region.[83,84] In addition, Liu et al reported that TOP2A gene expression was regulated by TP53 gene status and that several TP53 mutants exhibited reduced suppression of TOP2A gene expression.[85] This regulation was explained as a consequence of TP53 interfering with NF-Y binding to the regulatory sequences of the TOP2A promoter. NF-Y, a ubiquitous transcription factor, recognizes and binds to inverted CCAAT boxes (ICBs). The decrease in NF-Y activity is correlated with the decrease in TOP2A transcriptional activity, transcript level, and expression.[86] Furthermore, a chromatin immunoprecipitation analysis showed that reduced recruitment of NF-Y to TOP2A gene regulatory regions decreased its transcription.[87,88]

GC boxes are common elements in promoters. In the TOP2A promoter, GC1 and GC2 flank ICB1 and ICB5, respectively. It has been reported that GC1 has a major role in the basal transcription of TOP2A, while GC2 functions in a more modulatory capacity.[89] Proteins bound to the GC2 element may act as repressors. The specificity proteins Sp1 and Sp3 have been implicated in the regulation of TOP2A transcription through binding to both GC1 and GC2. Sp1 is commonly known as a transcriptional activator and is able to up-regulate transcription in a variety of promoters, including TOP2A, while Sp3 is bifunctional. Sp3 is a transcriptional repressor of TOP2A and a common modulator of Sp1-dependent transcriptional activation.[89] Since Sp3 is functionally dominant over Sp1, relatively small increases in Sp3 levels or decreases in the Sp1/Sp3 ratio could result in reduced transcription of TOP2A. However, Mo et al found that Sp3 was a transcriptional activator.[90] It is possible that this discrepancy is cell type specific.

POST-TRANSLATIONAL MODIFICATIONS OF TOP2A

In mammalian cells, several different mechanisms regulate TOP2A's decatenation ability. Posttranslational modification, including phosphorylation, ubiquitination, and small ubiquitin-like modifiers (SUMOylation), can alter its protein activity, stability, or localization (Figure 4C). The proteins involved in these modifications can regulate TOP2A's decatenation of intertwined daughter chromosomes. Mutations or alterations of the genes involving in TOP2A post-translational modification can lead to CIN and ultimately, cancer.

Phosphorylation

Phosphorylation is a key mechanism that regulates TOP2A's function. It has been proposed to affect the catalytic activity of TOP2A.[91] Most of the phosphorylation sites in TOP2A protein are located within the C-terminal domain (Figure 4A).[92-94] Some studies have reported that these phosphorylation sites are not important for enzymatic activity but are critical for the nuclear localization of TOP2A,[95] because deletion of this region or mutation of Ser-1376 or -1524 did not lead to inactivation of the enzyme.[95,96] However, Chikamori et al demonstrated that Ser-1106 was a major phosphorylation site in the catalytic domain of TOP2A and that mutation of this Ser-1106 to alanine caused a decrease in enzymatic activity.[97] In addition, hypophosphorylation of Ser-1106 may be correlated with etoposide resistance. These findings establish that Ser-1106 phosphorylation plays a critical role in regulating TOP2A's decatenation function.

Several proteins are involved in the phosphorylation of TOP2A. Casein kinase IIβ, protein kinase C, and extracellular signal-related kinase 2 have been shown to phosphorylate TOP2A and enhance its activity.[95,98,99] P38γ is a member of the p38 mitogen-activated protein kinase family that can be activated by both stress and mitogenic signals. Qi et al reported that p38γ phosphorylated the TOP2A Ser-1542 site and was important for TOP2A's stability and activity.[100] The polo-like kinase (Plk) family is also involved in the phosphorylation of TOP2A. Plk1 phosphorylates Ser1337 and Ser1524 of TOP2A, while Plk3 phosphorylates Thr1342.[101,102] The maximum level of phosphorylation occurs in mitosis, and Plk1- and Plk2-associated phosphorylation can activate TOP2A. Although phosphorylation does not affect TOP2A's dynamic localization in chromosomes, it is required for its essential role in sister chromosome segregation, and the observed hyperphosphorylation could represent compensation for the reduced protein level.[103]

Ubiquitination

As a tumor suppressor, BRCA1 plays a versatile role through its ability to participate in DNA damage response, checkpoint control, mitotic spindle assembly, centrosome duplication, and sister chromosome segregation[104]. BRCA1 plays a critical role in the maintenance of chromosome stability, which participates in TOP2A -dependent DNA decatenation. Cells derived from conditional BRCA1-knockout mice have various chromosome abnormalities.[105,106]

BRCA1 may regulate DNA decatenation through the ubiquitination of TOP2A. Lou et al reported that, in the absence of BRCA1, there was only one-third as much decatenation activity, and this contributed to a defect in chromosome segregation[107] that was also reported in BRCA1-defective embryonic mouse fibroblasts. BRCA1-deficient cells had many spontaneous chromosome abnormalities.[105,106,108] Shinagawa et al found that BRCA1 is a regulator of TOP2A ubiquitination.[109] Other studies have supported this mechanism and demonstrated that BRCA1 has a RING finger domain with documented E3 ubiquitin ligase activity.[107,110] BRCA1 can form a complex that includes BRCA1 and pRb and then ubiquitinate TOP2A for its degradation. In this complex, pRb is another key factor required for the degradation and inactivation of TOP2A on exposure to oxidative stress.

SUMOylation

SUMOs may be involved in diverse biological functions. The results of both genetic and biochemical studies indicate that the SUMO modification pathway plays an important role in proper cell cycle control, especially in the normal progression of mitosis. The results of studies in XEEs have implicated SUMOylation in TOP2A's decatenation ability.[111,112] TOP2A is modified by SUMO-2/3 on mitotic chromosomes in the early stages of mitosis (Figure 4A).[113] The inhibition of mitotic SUMOylation in the XEE assay system causes aberrant sister chromosome separation in anaphase and alters TOP2A's association with chromosomes.[114] Similarly, PIASγ, a member of the PIAS family, is unique in its capacity to bind mitotic chromosomes and is indispensable for accurate chromosome segregation in XEEs. PIASγ with SUMO E3 ligase activity is a critical regulator of the mitotic SUMO-2 conjugation of TOP2A,[115] and the activity of SUMO E3 ligase led to SUMO's modification of TOP2A in mitosis.[116] In human cells, PIASγ directs TOP2A to specific chromosome regions that require efficient removal of DNA catenations prior to anaphase. A lack of PIASγ leads to a prolonged metaphase block, in which normal metaphase plates form and the spindle checkpoint is activated. Moreover, sister chromatids remain cohered, even if cohesin is removed by depleting hSgo1, because DNA catenations persist at centromeres.[117]

Recently, it was reported that the nuclear pole protein RanBP2 bound to TOP2A and regulated its SUMO modification, specifically in mitosis. RanBP2 catalyzed the SUMOylation of TOP2A, directing this protein to inner centromeres for accurate chromosome separation prior to anaphase onset. When RanBP2 was insufficient, this defect led to chromatin bridges in anaphase, which were linked to impaired TOP2A-mediated decatenation of sister chromatids.[37] The results of these studies demonstrate that SUMO modification leads TOP2A to accumulate at inner centromeres and is essential for proper sister chromosome separation in mitosis.

Deacetylation

Histone acetylases (HATs) and histone deacetylases (HDACs) modify nucleosomal histones. Tsai et al reported that two histone deacetylation enzymes, HDAC1 and HDAC2, were associated with TOP2 in vivo under normal physiological conditions.[118] Their results indicate that TOP2A and TOP2B are substrates for HDAC1 and HDAC2 and that complexes containing HDAC1 or HDAC2 can increase TOP2 activity. HDAC1 and HDAC2 may facilitate chromatin modification by targeting a subunit of TOP2 to chromatin regions that are actively undergoing histone deacetylation. However, the mechanism by which TOP2 is modified by deacetylation in vivo requires further investigation. Although the balance between protein acetylation and deacetylation controls several physiological and pathological cellular processes, the role of HATs in acetylation of TOP2A is yet-to-be defined.

REGULATION OF TOP2A's DNA BINDING

Chromatin-Accessibility Complex

TOP2A is an ATP-dependent enzyme, and ATPase activity, which is important in TOP2A catalyzed hydrolysis, is stimulated by free DNA. The chromatin-accessibility complex uses energy to increase the general accessibility of DNA in chromatin. When its ATPase activity is stimulated by nucleosomal DNA, it facilitates TOP2A's hydrolysis ability. The ATPase subunit of the chromatin-accessibility complex, ISWI, and TOP2A are located in the same complex. Nucleosome remodeling by ISWI may help TOP2A bind to chromatin, which will enhanceTOP2A's hydrolysis.[119]

SWI/SNF Complex

A similar mechanism is found in another ATP-dependent chromatin remodeling complex, SWI/SNF. The mammalian SWI/SNF complex is composed of two distinct groups, the Brm/Brg1-associated factor (BAF) and polybromo-associated BAF (PBAF) complexes. The BAF complex is composed of one of two mutually exclusive catalytic ATPase subunits, BRG1 or BRM, whereas the PBAF complex uses only BRG1 as the catalytic subunit. The deletion of BRG1 or the expression of the tumor-associated G1232D mutant BRG1 reduced TOP2A's decatenation ability and caused anaphase bridges.[39] The loss of another subunit of the BAF complex, BAF250a, also resulted in TOP2A decatenation defects. In the BAF complex, TOP2A associated with BRG1 through a direct interaction with BAF250a. Moreover, TOP2A binding to DNA was dependent on BAF's function. The BRG1 mutants reduced the association between TOP2A and chromatin. Consequently, reduced binding of TOP2A to chromatin reduced TOP2A's function and weakened its ability to associate with substrate DNA during decatenation.

DNA Damage Checkpoint 1 Complex

A number of gene products are involved in TOP2A's decatenation function (Table 2). Mediator of DNA damage checkpoint 1 (MDC1) is an important multi-functional checkpoint protein that can interact with TOP2A.[96] Luo et al found within the 71 amino acids of the TOP2A C-terminal region (residues 1461-1531), Ser 1524 was required for MDC1 binding to TOP2A (Figure 4A); mutation of the Ser 1524 site abolished this interaction. In addition, phosphorylation of Ser 1524 occurred preferentially when cells were in G2/M transition.[96] These findings indicate that MDC1-TOP2A interaction and TOP2A Ser 1524 phosphorylation of are involved in the decatenation checkpoint. Luo et al further demonstrated that when phosphorylated, Ser 1524 of TOP2A acted as a binding site for MDC1 and the following MDC1-TOP2A interaction recruited this complex to chromatin. Although the MDC1-TOP2A interaction is not required for checkpoint activation induced by DNA damage, it is required for activation of the decatenation checkpoint. Mutation of Ser 1524 results in a defective decatenation checkpoint. These results reveal an important role of MDC1-TOP2A interaction in checkpoint activation and the maintenance of chromosome stability.

Table 2

Factors involved in regulation of TOP2A's activity

Factor	Description	Mechanism	Function	Ref.
CKIIβ	Casein kinase Iiβ	Phosphorylation	Enhance TOP2A activity	94
PKC	Protein kinase C	Phosphorylation	Enhance TOP2A activity	97
ERK2	Extracellular signal-related kinase 2	Phosphorylation	Enhance TOP2A activity	98
p38γ	p38 MAPK family	Phosphorylation of Ser1542 of TOP2A	TOP2A stability and activity	99
Plk1	Polo-like kinase 1	Phosphorylation of Ser1337 and Ser1524 of TOP2A	Activate TOP2A	100
Plk3	Polo-like kinase 3	Phosphorylation of Thr1342 of TOP2A	Activate TOP2A	101
BRCA1	E3 ubiquitin ligase activity	Ubiquitination	Reduce TOP2A activity	105
Rb	Retinoblastoma protein	Unknown	Inhibit TOP2A activity	111
SUMO2/3	SUMO isoforms	SUMOylation	Centromeric localization of TOP2A	114
PIASγ	SUMO E3 ligase	SUMOylation	Centromeric localization of TOP2A	114,115
RanBP2	Nuclear pore complex protein with SUMO E3 ligase activity	SUMOylation	Direct TOP2A to inner centromeres, accumulate TOP2A in metaphase	37
HDAC1/2	Histone deacetylase 1 and 2	Deacetylation	Enhance TOP2A activity	116
ISWI	Nucleosome remodeling factor	Interaction of TOP2A with chromatin	Facilitate the hydrolysis ability of TOP2A	119
BRG1	Transcription activator, member of the SWI/SNF family	Interaction of TOP2A with chromatin	Enhance TOP2A activity	39
MDC1	Multi-functional checkpoint protein	Unknown	Activate the decatenation checkpoint controlled by TOP2A	95

ALTERATIONS OF TOP2A AND RELATED GENES IN CANCERS TOP2A

TOP2A is considered a specific marker for cell proliferation, and it plays an important role in malignant tumors. As indicated by the findings discussed above, several interacting partners of TOP2A and mitotic checkpoint proteins regulate accurate chromosome segregation. The abnormal alteration of TOP2A and related genes consistently contributes to the human cancer genome (Table 3). The expression, genetic alteration, and enzyme activity of TOP2A have been studied in several types of malignancies.[120-125] In primary invasive ovarian carcinoma, both protein and mRNA levels of TOP2A expression were increased in high-grade and advanced-stage tumors; there was also a correlation between high expression and poor survival.[121] In Depowski et al's study, TOP2A was a candidate marker of increased cell proliferation and poor prognosis in breast cancer.[122] These authors also reported that it was preferentially expressed in a more aggressive subset of breast tumors (HER-2/neu overexpressed), and HER-2/neu was overexpressed in tumors with increased TOP2A expression. Since TOP2A is located close to HER-2/neu on chromosome 17, the levels of TOP2A gene copy number were assessed in samples with HER-2/neu amplification. Bhargava et al’ reported that TOP2A amplification was found in 39% (25/64) of HER-2-amplified tumors and was not detected in the absence of HER-2 amplification in their study.[154] TOP2A deletion was also seen in 7 (11%) of 64 tumors. However, whether TOP2A amplification leads to higher TOP2A protein levels is controversial. Mueller et al reported that TOP2A gene amplification in breast tumors was not predictive of high TOP2A protein expression.[123]

Table 3

Genetic aberrations of TOP2A and related genes in human cancers

Protein	Gene alteration event	Cancer type	Ref.
TOP2A	Amplified or deleted (depending on cancer types)	Breast cancer, esophageal cancer, ovarian cancer, gallbladder cancer, oral cancer, hepatocellular cancer, prostate cancer	120-125
P53	Mutation, deletion, epigenetic silencing	Lung cancer, breast cancer, gastric cancer, colorectal cancer, bladder cancer, ovarian cancer, hepatocellular cancer, head and neck cancer, pancreatic cancer, nasopharyngeal carcinoma, lymphoma	126-128
Rb	Mutation, deletion, epigenetic silencing	Lung cancer, breast cancer, gastric cancer, colorectal cancer, bladder cancer, ovarian cancer, hepatocellular cancer, head and neck cancer, pancreatic cancer, nasopharyngeal carcinoma, lymphoma	126-128
Sp1	Overexpression	Breast carcinomas, thyroid cancer, hepatocellular carcinomas, pancreatic cancer, colorectal cancer, gastric cancer, lung cancer, cervical cancer	129,130
HMGB1	Overexpression	Gastric colorectal cancer, prostate cancer, skin cancer, pancreatic tumor, NSCLC, hepatocellular cancer, head and neck cancer, bladder cancer, nasopharyngeal carcinoma	131-135
Plk1	Mutation, overexpression	Melanoma, breast cancer, ovarian cancer, thyroid cancer, colorectal cancer, prostate cancer, pancreatic cancer, head and neck cancer, NSCLC, non-Hodgkin lymphoma	136
PKC	Mutation, up- or down-regulation (depending on cancer type)	Gastric cancer, esophageal cancer, colorectal cancer, bladder cancer, breast cancer, ovarian cancer, hepatocellular cancer, basal cell cancer, prostate cancer, endometrial cancer, pancreatic cancer, renal cancer, B-cell lymphoma	137
p38γ	Overexpression	Breast cancer	138
BRCA1	Downregulation, mutation	Breast cancer, ovarian cancer, prostate cancer, pancreatic cancer, lung cancer	99, 139-145
BRG1	Mutation	Breast cancer, ovarian cancer, prostate cancer, lung cancer, pancreatic cancer, colorectal cancer	146-151
MDC1	Overexpression	Cervical cancer, breast cancer	152,153

In addition to human breast and ovarian cancer, elevated expression of TOP2A has been identified in oral cancer, nasopharyngeal carcinoma, esophageal cancer, lung cancer, gallbladder carcinoma, hepatocellular cancer, and colorectal cancer.[125] TOP2A overexpression in these cancers is associated with an aggressive tumor phenotype, advanced disease stage, tumor recurrence, and decreased overall survival.

BRCA1

As one of the representative interacting partners of TOP2A, BRCA1 is located on chromosome 17q21.31 and encodes a critical tumor suppressor. Low expression of BRCA1 was previously shown to increase the growth rate of benign and malignant breast tissue.[141] In another study, loss of nuclear BRCA1 expression was significantly associated with high histological grade.[142] Interestingly, low BRCA1 expression is associated with the development of distant metastasis and a poor prognosis in sporadic breast cancer.[143] Dorairaj et al reported that the rs8176318 G > T 3’UTR variant of BRCA1 is associated with decreased BRCA1 expression, both in vitro and in vivo. Moreover, they showed that this variant was predictive of aggressive and stage IV disease.[144]

Importantly, BRCA1 mutations account for the majority of hereditary breast and ovarian cancers. Compared to BRCA2-mutation carriers, BRCA1-mutation carriers have a higher risk of developing both breast and ovarian cancer. The risk of breast cancer in individuals with the BRCA1 mutation range from 50% to 80%.[145] Meanwhile, over 10% of human ovarian cancers contain BRCA1 mutations in the coding region, which result in genetic instability.[103] In addition to human breast and ovarian cancer, BRCA1 mutation carriers were found to have an increased risk of pancreatic cancer and prostate carcinoma.[140]

BRG1

BRG1 is a critical interacting partner of TOP2A and is mutated in various malignancies, including prostate, lung, breast, colon, pancreas, ovary, and colon carcinomas. Allelotype analyses have often shown loss of heterozygosity at the 19p13 region, where Brg1 is localized. In addition, a number of anti-cancer proteins, such as BRCA1, p53, and Rb, have been functionally linked to BRG1, further suggesting that Brg1 loss can hasten cancer development. The results of these studies suggest that BRG1 has a tumor-suppressive role in a wide range of human cancers.[146]

Medina et al reported a high incidence of inactivating mutations of Brg1 in lung cancer cell lines.[147] BRG1 was inactive in a quarter of the 59 lines. Such alterations were more common in non-small cell lung cancer (NSCLC) cell lines. Thirty-five percent of the NSCLC cell lines carried Brg1-inactivating mutations, compared with only approximately 5% of SCLC cell lines. An earlier study performed an immunohistochemical analysis to evaluate the levels of BRG1 in primary NSCLCs and found that 30% of tumors had no detectable protein, suggesting BRG1 inactivation.[148] Furthermore, loss of nuclear expression of Brg1 was associated with a low survival rate in NSCLC patients. Glaros et al found that heterozygous loss of Brg1 could serve as an initiating event in lung cancer development, whereas complete loss of Brg1 could promote tumor progression rather than tumor initiation.[149]

In pancreatic tumors, Brg1 also acts as a tumor suppressor. Mutations in Brg1 and other members of the SWI/SNF complex were observed in over 30% of human pancreatic ductal adenocarcinoma cases.[150] Furthermore, decreased Brg1 expression was associated with the IPMN precursor lesion that pancreatic ductal adenocarcinoma is thought to arise from.[151] As in lung cancer, Brg1-inactivating mutations and deletions were found in pancreatic cancer cell lines. Taken as together, loss of BRG1 affects cancer development.

THE POTENTIAL ROLE OF TOP2A IN PERSONALIZED CANCER THERAPY

The homeostasis and balance of the topoisomerase network are critical. Too much or too few of the key enzymes will lead to topological stress on the chromosomes and cell death. Thus, TOP2 and TOP1 are often highly expressed to cope with stress. This may lead to a “topoisomerase addiction” phenotype in CIN. Targeting this addiction may kill these cells.[155] In addition, the level of TOP2A increases 2-3-fold during G2/M phase,[156] and the expression level is much higher in rapidly proliferating cells than in quiescent cell populations.[157] Targeting both TOP2A-mediated DNA cleavage and cell proliferation has been an attractive approach in cancer therapy. Another isoform of TOP2, TOP2B, does not change significantly during the cell cycle. It has been suggested that targeting TOP2B leads to several undesirable consequences, with little clear benefit.[158]

TOP2-targeting Agents in Clinic

Several TOP2 inhibitors/poisons have been approved by the US Food and Drug Administration, including etoposide, teniposide, doxorubicin, idarubicin, epirubicin, and mitoxantrone (Figure 5). Both etoposide and teniposide damage DNA by interacting with TOP2 to form cleavable complexes that prevent the relegation of DNA; this leads to double-strand DNA breaks. Etoposide is used to treat a wide variety of malignancies, including SCLC, testicular cancer, neuroblastoma, leukemias, and lymphomas, while teniposide is used in pediatric patients with poor-prognosis acute lymphocytic leukemia.[157] Like etoposide and teniposide, anthracyclines intercalate into DNA and are commonly used to treat breast cancer, leukemias, lymphomas, and sarcomas. The anthracyclines that are currently approved for use in the United States are doxorubicin, daunorubicin, epirubicin, and idarubicin. Doxorubicin is the most commonly used anthracycline for the treatment of solid tumors. Mitoxantrone is an anthracenedione that targets TOP2 and is the only agent of its class approved for clinical use. In the past, these drugs were used without considering whether the cancer had a G2/M checkpoint defect and CIN. Now, an important goal of clinical research is to maximize the therapeutic efficacy of TOP2-targeting drugs while minimizing the risk of toxicity. Thus, it is critical to better understand the role of TOP2A and the genes that regulate TOP2A in personalized chemotherapy.

Figure 5

TOP2-targeting agents in chemotherapy

Drugs that target TOP2 fall into one of two categories: poisons (red) or catalytic inhibitors (blue). US Food and Drug Administration-approved TOP2-targeted inhibitors (grey) include etoposide, teniposide, doxorubicin, idarubicin, epirubicin, and mitoxantrone.

Potential Roles of TOP2A in Personalized Cancer Therapy

The TOP2A level is believed to be a major determinant of cellular sensitivity to targeting this enzyme. Thus, genetic and molecular defects that lead to increased TOP2A levels may serve as predictors of a better response to TOP2A inhibitors/poisons. Because TP53 deficiency is one of the most common genetic aberrations in human cancers and TP53 negatively regulates TOP2A expression, cancer cells with the TP53 mutation have elevated TOP2A levels and are more sensitive to TOP2A inhibitors/poisons.[85] Similarly, it has been shown that the level of TOP2A partly depends on the ratio of Sp1:Sp3 in the cells, which also affects the sensitivity of cancer cells to TOP2A inhibitors.[89] Wang et al reported that alterations in NF-Y, a ubiquitous transcription factor that interacts with the TOP2A promoter, had a significant effect on TOP2A expression and has important consequences in TOP2A-targeting agents.[86] Identifying the key factors that regulate TOP2A will help us develop an accurate model for predicting patients’ response to TOP2A-targeted chemotherapy.

Because TOP2A is an enzyme, its activity may more closely determine cellular sensitivity to TOP2A-directed chemotherapy. Cancer cells with a defect in TOP2A decatenation activity, which is caused by TOP2A modifications or other factors, are resistant to topoisomerase inhibitors/poisons. Since phosphorylation of TOP2A regulates its decatenation activity, researchers have studied the role of altered TOP2A phosphorylation in drug sensitivity. Chikamori found that cancer cells with the TOP2A S1106A site mutation were resistant to etoposide and amsacrine,[97] demonstrating that site-specific phosphorylation regulates sensitivity to TOP2-targeting drugs. Thus, hypophosphorylation of TOP2A may be responsible for patients’ lack of response to treatment. Chen et al reported that hypophosphorylation of TOP2A in teniposide-resistant cells was two times higher than in parental cells.[159] A subsequent study by Ritke et al in etoposide-resistant K562 human leukemia cells revealed that hypophosphorylation of TOP2A in these cells was due to decreased levels of Casein kinase IIβ.[160] In addition, as it is a kinase, ERK2 has been shown to phosphorylate TOP2A and enhance its activity.[99] Kolb reported that inhibition of ERK1 and ERK2 activation using specific inhibitors markedly attenuated the G2/M arrest induced by etoposide.[161] Similarly, p38γ, which is important to TOP2A's stability and activity, actively regulates the drug-TOP2A signal transduction; this can be exploited to increase the therapeutic activity of TOP2 drugs. In addition to phosphorylation, other posttranslational modifications of TOP2A, including ubiquitination, can regulate the decatenation and localization of this enzyme. Unlike phosphorylation, which can enhance TOP2A activity, BRCA1 ubiquitination reduces TOP2A activity.[102] The absence of functional BRCA1 enhances cellular sensitivity to etoposide.[162]

A major challenge to the clinical use of TOP2 inhibitors/poisons is that they are associated with the development of secondary cancers. Etoposide is widely used for the treatment of many cancer types; however, it is associated with an increased risk of secondary leukemia, particularly acute myelogenous leukemia. TOP2 inhibitors/poisons, including etoposide, often cause rearrangements that involve the mixed lineage leukemia gene on chromosome 11q23, which is associated with this secondary cancer. The risk of secondary acute myelogenous leukemia appears to be dependent on the drug dose. Pedersen-Bjergaard reported that the risk of developing secondary leukemia was 336 times higher with etoposide doses of > 2.0 g/m2 than with doses of ≤ 2.0 g/m2.[163] On the basis of these data, CIN status may be an important factor that should be considered when the dose of TOP2 inhibitors/poisons are chosen because cancer cells with different TOP2A decatenation activity levels have different sensitivity to topoisomerase inhibitors and low-dose regimens induce a lower risk of leukemogenesis.

Another challenge in clinically using TOP2-targeting agents is drug resistance. In resistant cells, TOP2A alterations have been found at both expression level and catalytic activity of the enzyme. Point mutations in the TOP2A gene have also been found in resistant cells. Cancer cells with the TOP2A decatenation defect are resistant to TOP2 inhibitors, indicating that cancer cells’ sensitivity to these drugs can be affected by genes that are involved in TOP2A-mediated CIN. Mutations or alterations in these genes may result in different prognoses after treatment with topoisomerase inhibitors. Thus, further studies of TOP2 drug sensitivity and CIN genes may open a new avenue for the development of personalized cancer therapy.

CONCLUDING REMARKS

A comprehensive characterization of cancer genome landscapes unequivocally established that genomic instability is a hallmark of tumorigenesis. Genomic instability manifests in two major genotypes: genetic mutations and CIN. Genetic mutations are believed to be initiated by mutations of a group of DNA repair genes. However, the initiating events for CIN are still elusive. The picture is further muddied because chromosomal copy number alterations are also observed in normal cells; thus, the initiating events for CIN may regularly occur in subclones of normal cells, but only a small percentage evolve into neoplasms.[164] Therefore, although CIN is fully manifested as a tumor phenomenon, it is unlikely to be tumor specific. Rather, it is mostly likely a pre-cancer phenotype that is intimately tied to cell division, a highly regulated and fundamental process with multiple checkpoints to monitor for errors and remove strayed cells. In particular to the CIN phenotype, correct chromosome segregation requires extensive choreography with key “conductors,” such as TOP2A to cut and paste, spin, and package. The more cell divisions that occur, the higher the chance of errors and the higher the chances of cells surviving the monitoring system and becoming neoplastic. Cancer cells either have CIN or the high mutation phenotype, illustrating that not only do these two genetic aberrations have distinct mechanisms but also that they have an equilibrium and “stability” that is likely supported by a new set of machinery. In other words, these CIN cells are “addicted” to something, which if characterized, would qualify as the Achilles heel.

The results of our literature review and analysis support the hypothesis that TOP2A is an Achilles heel in cancer therapy. However, the reported responses to TOP2 inhibitors/poisons have been limited. This is not surprising given that not all cancer cells are “addicted” to TOP2. TOP2's activities are subject to multiple levels of regulation—copy number, gene expression control, post-translational modifications, partner protein interactions, and feedback loops—that lead to other changes in checkpoint control, such as the compensatory regulation of TOP1. Therefore, to fully explore TOP2 as an effective target for therapy, more in-depth investigations of the TOP2 regulatory network are needed. Careful interrogation of the clinical annotations that are tied to genomic characterization will help us identify prognostic factors that can be used to stratify patients by responsive or resistant disease; only those with responsive would receive TOP2 inhibitors/poisons.