Introduction

As one of the major components of the cytoskeleton, microtubules are key players in cellular processes such as cell migration, intracellular trafficking, and DNA segregation during mitosis [1, 2]. Microtubules are assembled from 13 laterally associating protofilaments, which in turn consist of α- and β-tubulin heterodimers aligned in a head-to-tail pattern (Fig. 1a). The alignment of α- and β-tubulin heterodimers leads to the polarity of the microtubule, with β-tubulin exposed at one end (plus end) and α-tubulin exposed at the other end (minus end) [1]. In most mammalian cells, the minus ends of microtubules are clustered at microtubule organizing centers, while the plus ends face the cell cortex [3]. Most microtubules in the cells are in consistent transitions between growth and shrinkage by associating and dissociating of α/β-tubulin heterodimers at both ends, which is known as dynamic instability (Fig. 1b) [4]. The highly dynamic properties of microtubules are essential for many cellular activities, particularly cell division [1]. Mounting evidence shows that dysregulation of microtubule dynamics contributes to the development of serious diseases, including cancer [5,6,7,8].

Fig. 1
figure 1

a A schematic illustration showing that microtubule dynamics are regulated by different types of MAPs, including canonical MAPs, microtubule plus-end-tracking proteins, centrosomal proteins, microtubule motor proteins, etc. The action of paclitaxel in the microtubule network occurs through binding to β-tubulin. b, c HeLa cells were stained with anti-tubulin antibody, and then with fluorescein-conjugated secondary antibodies. b Microtubules are highly dynamic structures and undergo frequent transitions between growth and shrinkage, a phenomenon known as dynamic instability. c Paclitaxel promotes microtubule assembly and disrupts microtubule dynamics

Full size image

Vinca alkaloids and taxanes are two families of anti-microtubule agents wildly used in clinical treatment of cancer including solid tumors and hematological malignancies [9]. Paclitaxel, a member of the taxane family, is a natural alkaloid originally isolated from the bark of a tree, Taxus brevifolia. The anti-tumoral activity of paclitaxel was discovered in the 1970s and it was first approved by the US FDA in 1992 for treatment of ovarian cancer. In addition, paclitaxel has proven to be effective in other solid tumors, including breast cancer and non-small-cell lung cancer [10, 11]. Preliminary mechanism showed that paclitaxel acts by stabilizing cellular microtubules and blocking chromosome segregation (Fig. 1c) [12]. Moreover, it is established that the action of paclitaxel in the microtubule network occurs through interaction with β-tubulin. In the following years, β-tubulin functional domains and the domains containing paclitaxel-binding sites were identified [13].

Despite the great success of paclitaxel, the variable response activity of patients to the drug limits its clinical utility. Acquired resistance to paclitaxel is one of the most significant reasons for its failure in chemotherapy [14]. Previous studies have demonstrated that paclitaxel resistance is attributable to various mechanisms: elevated drug efflux that results from upregulation of membrane transporters such as P-glycoprotein [15, 16]; alterations in the expression of β-tubulin isotypes [17]; and changes in apoptotic regulatory proteins such as Bcl-2 [18,19,20]. In spite of decades of great efforts worldwide, the precise mechanisms underlying paclitaxel activity remain largely unknown. In this review, we will focus on the roles of MAPs (microtubule-associated proteins) in regulation of paclitaxel activity in different types of cancers. Elucidation of these issues might provide significant diagnostic or prognostic value, as well as potential targets for cancer chemotherapy. We categorize MAPs into canonical MAPs, plus-end-binding proteins, centrosomal proteins, and other proteins that might affect the properties and functions of microtubule, and detail their roles in the regulation of paclitaxel activity.

Distinct roles of MAPs in paclitaxel activity regulation

MAPs represent a large number of proteins that act in concert with the microtubule network, and regulate microtubule properties and functions [21,22,23,24]. Canonical MAPs are structural MAPs, such as MAP2 and tau. MAPs regulate microtubule behavior by binding microtubules either at the microtubule ends or on the outside of the microtubule wall or interacting with the soluble tubulin pool (Fig. 1c) [25, 26]. MAPs participate in various microtubule-mediated cellular activities by modulating microtubule behavior such as stability, assembly or bundling [21, 26, 27]. Accumulating evidence shows that changes in the expression or post-translational modification of MAPs lead to the alteration of microtubule dynamics, and are associated with the development of serious diseases including cancer [5, 6, 28, 29]. The dysregulation of microtubule dynamics by MAPs may result in tumor resistance to chemotherapeutic agents that target microtubules, and many MAPs with aberrant expression have been shown to be correlated with susceptibility to paclitaxel [30, 31].

Tau, a microtubule-associated protein, promotes tubulin polymerization and stabilizes microtubules [32]. It is well known that tau plays an essential role in the regulation of nervous system, and is significantly involved in the pathogenesis of Alzheimer’s disease as well as the tauopathies [33]. Overphosphorylation of tau, mainly occurring in the axons, contributes to neurofibrillary degeneration and is associated with cell dysfunction and death. Tau binds to β-tubulin in the same site as paclitaxel, and consequently competes with the drug. An in vitro study shows that preincubation of tubulin with Tau decreases paclitaxel binding and reduces paclitaxel-induced microtubule polymerization in breast cancer cell line [30, 34]. In addition, high expression of Tau shows a significant association with poor response to paclitaxel chemotherapy in patients with metastatic breast cancer [35]. Moreover, reduction of Tau expression sensitizes ovarian carcinoma to the paclitaxel treatment [36]. These studies show that Tau expression may act as a predictor of response to paclitaxel in cancer patients.

MAP2 overexpression has been observed in several types of cancers [37,38,39,40,41]. Similar to tau, MAP2 also binds to and stabilizes microtubules. However, they regulate paclitaxel activity of cancer cells via different mechanisms. Increased expression of MAP2 in breast cancer cell lines leads to increased paclitaxel sensitivity [42]. Gene expression analysis shows that the expression level of MAP2 is significantly higher in breast cancer patients achieving a pathologic complete response to neoadjuvant paclitaxel with radiation, indicating that patients with elevated MAP2 expression might show a better response to paclitaxel treatment. Moreover, MAP7 domain-containing protein 3 (Mdp3), a microtubule binding protein, has been shown to be involved in breast cancer growth and metastasis [7]. Besides, Mdp3 promotes microtubule assembly and stability by interacting with tubulin and microtubules [43]. It will be interesting to investigate the role of Mdp3 in paclitaxel activity modulation in cancers, which might be a novel predicting target in cancer chemotherapy.

Microtubule plus-end-binding proteins that promote paclitaxel sensitivity

Microtubule plus-end-binding proteins (+TIPs) localize to the growing plus ends of microtubules and couple microtubules to cellular structures, such as chromosomes and cell cortex [44, 45]. Microtubule plus-end-binding proteins regulate microtubule dynamics and functions. Amounting studies have emphasized the central role of end-binding family proteins (EB) among the +TIPs [45,46,47,48,49]. Mammals contain three EBs (EB1-3), sharing 57–66% sequence identity, are encoded by various genes [50]. EB proteins bind directly to the growing microtubule plus ends via their N-terminal calponin homology domain, whereas their C terminus is important for associating with other +TIPs [51]. Among EB family proteins, EB1 and EB3 promote microtubule assembly and stabilization [52]. Interestingly, EB1 and EB3 sensitize microtubules to microtubule-targeting agents by regulation of microtubule dynamics [53,54,55].

It is recently reported that the expression of EB1 in breast tumor tissues correlates with the clinical response of the patients to paclitaxel-based chemotherapy [56]. Mechanistic study has showed that EB1 increases paclitaxel activity to cause mitotic arrest and apoptosis in breast cancer cell lines. In addition, EB1 stimulates paclitaxel sensitivity by promoting paclitaxel-microtubule association, thus increasing the activity of paclitaxel to enhance microtubule assembly and stabilization. It is also proved in other studies that EB1 is crucial for regulating the sensitivity of breast cancer cells to paclitaxel, through a mechanism in which it modulates paclitaxel-mediated stabilization of microtubule polymerization and regulates the binding sites of paclitaxel at the growing microtubule ends [57].

Cytoplasmic linker protein 170 (CLIP-170) is another member of the +TIPs family that binds to microtubule plus ends in an EB1-dependent manner [58]. CLIP-170 is regarded as a mediator of paclitaxel sensitivity in breast cancers [31]. The expression of CLIP-170 in breast cancer samples correlates with the pathological response of tumors to paclitaxel-containing chemotherapy. Similar to EB1, CLIP-170 regulates paclitaxel sensitivity in breast cancer cells by modulating the effects of paclitaxel on microtubule assembly, thus affecting cell-cycle retention and apoptosis.

Centrosomal proteins that modulate paclitaxel resistance

Centrosomes, the microtubule organizing centers, nucleate and organize microtubules in animal cells [59]. Centrosome is critical for cell cycle progression, by organizing radial microtubule array in interphase and establishing bipolar spindle during mitosis [3, 60, 61]. Centrosomes undergo maturation and segregation during the cell cycle, and centrosome abnormalities, such as supernumerary centrioles, abnormal centrosome number or volume, and altered expression of centrosomal proteins, have been implicated in tumorigenesis [62,63,64,65,66]. Overall, centrosome aberrations are positively associated with advanced cancer grade and metastasis [67]. It is implied that centrosome abnormalities may have clinical diagnostic and/or prognostic value and centrosomes may also be a potential target for cancer therapy. The correlation between centrosome protein expression and sensitivity to paclitaxel has attracted increasing attention in the field [68,69,70].

Ninein-like protein (Nlp), an important centrosomal protein involved in centrosome maturation, microtubule nucleation, and spindle formation, plays an essential role in tumorigenesis [71]. Nlp is found to be overexpressed in human breast, lung and ovarian carcinomas [72,73,74]. A study has investigated the correlation between Nlp and paclitaxel chemosensitivity, and found that overexpression of Nlp confers breast carcinoma resistance to paclitaxel [72]. By suppressing paclitaxel-induced microtubule polymerization, Nlp overexpression promotes mitotic arrest and suppresses apoptosis elicited by paclitaxel. Moreover, the expression of Nlp in breast cancer patients is highly correlated with their sensitivity to paclitaxel, and patients with high expression of Nlp are likely resistant to the treatment of paclitaxel.

The mammalian transforming acidic coiled coil (TACC) family of centrosomal proteins consists of three members (TACC1-3), which act as important structural components of the mitotic spindle apparatus [75]. TACC proteins are evolutionarily conserved and modulate centrosome integrity, microtubule assembly, and mitotic spindle stability [76,77,78,79]. Mounting evidences indicate that TACCs are involved in the progression of some human malignancies [80]. The expression of TACC1 has been shown to be upregulated in human cancer cell lines, and it is associated with mammary tumorigenesis and gastric carcinoma [81, 82]. TACC2 plays an important role in breast cancer cell proliferation, and its immunohistochemical status might be a candidate of worse prognostic factor in breast cancer cases [83]. Abnormal expression of TACC3 is associated with the etiology of ovarian, bladder and non-small-cell lung cancer [84,85,86]. Knockdown of TACC3 leads to breast cancer cell arrest in G1 through cellular senescence by increasing nuclear p21WAF and reducing retinoblastoma protein (Rb) and extracellular signal-regulated kinase 1/2 (ERK1/2) [87]. Treatment of breast cancer cells with a low concentration of paclitaxel promotes cellular senescence triggered upon TACC3 depletion, showing that blocking of TACC3 regulates paclitaxel sensitivity by modulating a premature senescence program.

Another centrosomal protein, Salt Inducible Kinase 2 (SIK2) has been identified as cell cycle regulator of ovarian cancer cells in a high-content siRNA kinome screen [88]. SIK2 plays a key role in the initiation of mitosis and regulates the localization of the centrosome linker protein, C-Nap1 [89, 90]. Depletion of SIK2 expression induces centrosome splitting in interphase, while it inhibits centrosome separation in mitosis, thus sensitizing ovarian cancers to paclitaxel both in ovarian cancer cells and xenografts. Compared to paclitaxel-sensitive ovarian cancers, the expression of SIK2 is significantly higher in paclitaxel-resistant cancers. These findings imply that SIK2 might be a therapeutic target in ovarian cancers. Centrosomal protein 70 (Cep70) is an important factor that mediates breast cancer growth and metastasis. The expression of Cep70 is upregulated in breast cancer tissues and cell lines, which is closely correlated with several clinicopathologic variables associated with breast cancer progression [8]. It has been shown that Cep70 participates in microtubule polymerization and stability, as well as mitotic spindle orientation [91,92,93,94]. Elucidation of the correlation of Cep70 expression with paclitaxel activity in breast cancer will improve the understanding of the role of centrosomal proteins in cancer chemotherapy [95].

Regulation of paclitaxel activity by other MAPs including spindle assembly checkpoint and microtubule motor proteins

Chromosome instability and aneuploidy are hallmarks of aggressive solid tumors [96, 97]. The spindle assembly checkpoint (SAC) has a crucial function in genetic integrity, which ensures accurate chromosome segregation during mitosis and prevents aneuploidy [98]. Impaired SAC function has been suggested as one of the causes of aneuploidy in human cancers [99]. SAC is a complex of proteins that includes Mad1, Mad2, Bub1, BubR1, Bub3, and MPS1 [100]. Among all SAC components, Mad2 and BubR1 have a pivotal function in checkpoint signaling due to their crucial function in the metaphase-to-anaphase transition by delaying anaphase onset, through inhibition of the anaphase promoting complex/cyclosome (APC/C) [101].

Recent studies have shown that paclitaxel sensitivity is associated with the spindle checkpoint. It has been reported that Mad2 downregulation induces premature senescence in breast cancer cells, and these senescence cells maintain viability, while replicative incompetent [101]. The paclitaxel treatment in Mad2-depleted cells shows a significant further induction of senescence compared with paclitaxel controls, indicating that compromised Mad2 levels lead to paclitaxel resistance of senescent breast cancer cells. Besides, Mad2 and BubR1 are significantly expressed in gastric cancer cells, and contribute to cellular transformation and tumorigenesis [102]. Similarly, paclitaxel treatment increased cell senescence in gastric cancer cells interfered for Mad2 or BubR1 expression. Intriguingly, it is found that reduced expression of BubR1, but not Mad2, is associated with acquired paclitaxel resistance in ovarian carcinoma cells [103].

Another type of key participants during mitosis is kinesin superfamily protein (KIF), which plays crucial roles in chromosomal and spindle movements [104]. KIFs are a conserved class of motor proteins that transport cargoes by walking unidirectionally along microtubule tracks, hydrolyzing one molecule of ATP at each step [105]. KIFs are implicated in a variety of cellular functions, such as mitosis, signal transduction, microtubule polymer dynamics, and intracellular transport [106]. Previous studies suggest that KIFs may play a key role in the development or progression in many types of cancers [107,108,109]. Kinesins have also gained attentions as potential targets for mitotic drugs. It is reported that overexpression of kinesin proteins, including KIFC3, KIF5A, and KIF12 contributes to paclitaxel resistance in BLBC (basal-like breast cancer) cell lines and tissues [110]. Moreover, the expression level of KIF14 is significantly increased in cervical cancer tissues, which is positively correlated with high-tumor stage, lymph node metastasis, paclitaxel chemoresistance as well as poor survival [111]. These findings indicate that KIF14 might serve as a novel prognostic biomarker of chemoresistance to paclitaxel treatment in cervical cancer (Table 1).

Table 1 Regulation of paclitaxel activity by MAPs
Full size table

Conclusion

In spite of the development of small-molecule pathway inhibitors, chemotherapy remains the treatment of primary choice for patients with cancer. Paclitaxel, a member of the taxane family, has gained some success in treating some solid tumors including breast cancer and non-small-cell lung cancer. However, the variable response activity of patients and acquired resistance to paclitaxel limit its clinical utility. It is outlined in the review that some MAPs promote paclitaxel sensitivity by regulating microtubule dynamics. Altering the expression of these proteins might promote the effects of paclitaxel on microtubule dynamics and contribute to a synergistic inhibition of cancer cell proliferation. Moreover, the expression level of several centrosomal proteins is upregulated in cancer, which confers resistance to the treatment of paclitaxel. In summary, these proteins might act as potential markers to predict response to paclitaxel, as well as be regulated to boost paclitaxel sensitivity. These findings not only provide theoretical evidence into more rational chemotherapeutic regimens in clinical practice but also novel insights into precision cancer treatment.