The role of PLK1 in cancer exhibiting chromosomal instability

ABSTRACT

Adenomatous polyposis coli (APC) mutations cause aneuploidy and are responsible for familial adenomatous polyposis characterized by chromosomal instability. PLK1 contributes to sustain an intact spindle assembly checkpoint ensuring genomic stability. In our work using independent Apc^Min/+ mouse models we revealed that PLK1 functions as tumor suppressor in APC-mutated colorectal cancers.

Colorectal cancer (CRC) is the third most diagnosed and second deadliest cancer amongst men and women. It is associated with a high genetic risk factor. Almost 5% of all CRC are ascribed to hereditary factors and can be split into familial adenomatous polyposis (FAP) and the hereditary nonpolyposis colorectal cancer. Germline mutations of the tumor suppressor gene adenomatous polyposis coli (APC), in particular carboxy-terminal truncations, are key and early events resulting in the development of FA.¹ Through its C-terminal domain, the APC protein binds to the microtubule array during mitosis and it is mainly enriched at the microtubules + ends, where it stabilizes the microtubule-kinetochore attachment.² Additionally, it has been reported that APC associates with the spindle assembly checkpoint (SAC)-proteins BUBR1 and BUB1, thus, attributing a putative role to APC during mitotic exit. Accordingly, the truncated APC gene is reported to cause defects in mitotic spindles and chromosome segregation, two events that enhance chromosomal instability (CIN) in CRC. It is widely accepted that a compromised SAC plays a significant role in the emergence of CI.³

Polo-like kinase 1 (PLK1) is a major regulator of mitotic progression and key target in human cancer therap.⁴ Based on the activity of PLK1 in regulating SAC proficienc,⁵ on its essential function during mitosis and on the fact that PLK1 is upregulated in a panel of human cancers, we were keen to explore the role of PLK1 in CRC cells expressing a truncated form of APC (amino acids 1–750), which is missing the carboxy-terminal portion of the APC protein, herein referred to as APC-∆C. To shed new light on the role of PLK1 in human cancer with genomic instability, which is the major driving force of tumorigenesis, we performed this study in two CRC cancer cell lines exhibiting different genetic backgrounds, HCT 116 (APC full length, CIN negative) and SW480 (truncated APC, CIN positive.⁶ In both cell lines, we stably expressed the truncated APC variant APC-∆C. Upon PLK1 inhibition using the selective inhibitor BI6727, we were surprised to observe a rather accelerated cell growth, an improved cell survival, which rendered APC-∆C cells more resistant against cells death induced by the PLK1 inhibitor BI6727 in relation to their parental counterparts. Real time microscopy revealed that in absence of PLK1 activity, APC-∆C cells mainly escaped mitotic arrest, a phenomenon known as “mitotic slippage,⁷ without DNA segregation leading to the formation of polyploid cells with a high degree of chromosomal instability. Concomitantly, the analysis of metaphase chromosome spreads revealed that under PLK1 inhibition APC-∆C cells exhibit up to twice as many chromosomes compared to the parental cells indicating the important genetic disorder prevailing in these cells. These observations strongly suggest that the combination of the APC-truncation and PLK1 inhibition works synergistically towards weakening the activity of SAC, which results in premature and improper mitotic exit. In accordance with this, we established that PLK1 inhibition in APC-∆C cells hampers the association and the kinetochore recruitment of SAC signaling proteins (BUBR1, MAD1/2, BUB3 and CDC20). This is likely happening directly through loss of Plk1 phosphorylation on KNL1 and also indirectly by displacing Aurora B, the kinase required for the spindle checkpoint signaling, from the kinetochore⁸ (Figure 1). However, another intriguing aspect emerged from our study as well. Considering the role of APC full length in stabilizing kinetochore-microtubule attachments, it seemed that the sole expression of APC-∆C, altered the structural configuration of kinetochores, as the recruitment of certain core components, such as KNL1, HEC1, ZWINT-1 varied in contrast to the parental counterparts. Yet, the molecular mechanisms behind this remain elusive and require suitable experiments allowing in-depth insights into kinetochore assembly and structure, such as in vitro reconstitution of kinetochore component in presence of full length APC and APC-∆C.

Figure 1.

Combined effect of APC-ΔC expression and PLK1 inhibition in colorectal cancers (CRC).

PLK1 inhibition in APC-ΔC-expressing cells lead to a partial kinetochore delocalization of Aurora B and to an impaired spindle assembly checkpoint (SAC) activity due to a reduced recruitment of its components BUBR1 and MAD2 to kinetochores. This causes mitotic slippage and subsequently rises aneuploidy in CRC.

Furthermore, we could clearly demonstrate for the first time in a transgenic mouse cancer model the importance of Plk1 in the intestinal neoplastic transformation. Two mouse models were used to address this question: an Apc^Min/+ model and our previously established murine model for an inducible knockdown of Plk1 (iKD.⁹ By means of in vitro fertilization mice carrying the double allele Apc ^Min/+Plk1^ikd were generated. In our mouse models to inhibit the function of Plk1, mice were treated with BI6727 or RNAi was induced. The subsequent histological examination of the intestinal tissues demonstrated that in both models that the inhibition/downregulation of Plk1 increased the number of adenomatous polyps and more importantly, these polyps displayed the specific characteristic of high-grade adenomas or adenocarcinomas. Our data mining initiative revealed that in colon cancer patients harboring a nonsense mutation causing APC truncation, the patient group exhibiting high PLK1 expression level had a significantly higher survival rate compared to the patient group with low PLK1 expression supporting our observations in the two independent APC mouse models.

Remarkably, APC inactivation seems also to play an important role in different tumors of epithelial origin including breast cancer. In this context, APC inactivation occurs through promoter hypermethylation and this epigenetic modification can be detected in all pathological grades and stages of this diseas.¹⁰ Hence, it may be very interesting to investigate the potential role of PLK1 in breast cancer showing an APC inactivation phenotype and to explore whether the level of PLK1 expression correlates with the prognosis of breast cancer patients harboring an inactivation of APC functions.

The data we collected during this study strongly argue for the importance of PLK1 in the suppression of intestinal cancers with chromosomal instability and go against the idea that PLK1 upregulation in tumorigenesis is merely caused by the increased cell proliferation rather than a functional deregulation during this process. More importantly, we could demonstrate that in colon cancers that harbor APC truncation and are genetically instable, PLK1 is an indispensable tumor suppressor factor safeguarding the accurate progression of mitosis and the cell division. Finally, this work also provides a rational to develop combinatorial treatment towards strengthening the spindle assembly checkpoint, which may likely engender promising results in aneuploid tumors.

Disclosure of Potential Conflicts of Interest

No potential conflict of interest was disclosed.