Page 55 - Read Online

Page 8 of 20 Jusino et al. J Cancer Metastasis Treat 2018;4:43 I http://dx.doi.org/10.20517/2394-4722.2018.24

Centrosome amplification is defined as an excess of normal components, specifically more than two
centrosomes and more than four centrioles . Centrosome amplification results in multipolar or
[138]
pseudobipolar mitotic spindles that may culminate in aneuploidy and chromosome instability . Also,
[101]
centrosome amplification may lead to defects in cytokinesis that lead to tetraploidy . Because tetraploidy
[139]
and excess chromosome instability are associated with decreased cellular fitness [140,141] , cells with amplified
centrosomes avoid cell death by clustering centrosomes in order to avoid the generation of multipolar
mitosis, and excessive aneuploidy and chromosome instability [142,143] . However, cells with pseudobipolar
spindles form merotelic attachments that lead to single chromosome gains and losses . Either tetraploidy
[144]
or single chromosome losses have been shown to be tumorigenic in mouse models of cancer [145,146] . In a
more recent study, Sabino et al. demonstrated that Drosophila melanogaster epithelial wing disc cells
[147]
overexpressing Sak display extra centrosomes and exhibited mechanisms of clustering, but also inactivation
of extra centrosomes. Inactivation of extra centrosomes is defined as the gradual loss of microtubule-
nucleating capacity. Although inactivation culminates in normal spindle bipolarization, neither clustering
nor inactivation was efficient and abnormal segregation was observed. Furthermore, epithelial cells with
extra centrosomes generated tumors when transplanted into the wild-type host.

Although the role of numerical aberrations (i.e., centrosome amplification) in cancer has been extensively
studied, its role in tumor initiation, progression, and metastasis remains controversial, and may be context-
dependent. For example, centrosome amplification in hepatobiliary cancer is not associated with tumor
stage, size or proliferative activity . Likewise, there is no significant relationship between centrosome
[94]
amplification and tumor size, stage or patient survival in lung cancer . Moreover, studies from the Cleveland
[94]
group in mice - with centrosome amplification induced by Cre-recombinase-mediated Plk4 expression - did
not result in spontaneous tumorigenesis regardless of p53 status . Concordantly, studies from the Basto’s
[148]
laboratory demonstrated that induction of centrosome amplification in mouse brains caused microcephaly
due to increased apoptosis caused by multipolar divisions of neuronal stem cells . Our own studies using
[149]
an orthotopic model of breast cancer showed that rescuing back centrosome amplification in Her2 breast
+
cancer cells silenced for E2F3 through the overexpression of GFP-Nek2 did not influence tumor growth
or tumor burden . In contrast, other models suggest that centrosome amplification can influence tumor
[150]
initiation and progression. For example, centrosome amplification correlates with poor prognostic factors
such as nodal status and hormone receptor-negative status in 103 primary invasive breast cancers .
[151]
Likewise, centrosome amplification is associated with triple-negative breast cancers, higher stage, and higher
grade, correlating with decreased overall survival and relapse-free survival in a cohort of 362 breast cancer
patients . Another study confirmed the above results and correlated centrosome amplification with markers
[152]
of aggressiveness in triple-negative breast cancer patients, including increased stage and the mesenchymal
marker vimentin . Several transgenic models suggest that centrosome amplification might have causal,
[153]
rather than consequential effects on cancer. For example, centrosome amplification causes tumors in flies
independently of chromosome instability [154,155] . Other studies using transgenic mouse models involved the
temporal expression of the prolyl isomerase Pin1 , Aurora A , or K-Ras G12D[25] in mammary epithelial
[157]
[156]
cells, which resulted in pre-malignant mammary epithelial lesions with centrosome amplification that
preceded mammary tumors. In mice, centrosome amplification induced by Plk4 accelerates the time of onset
of lymphomas and sarcomas associated with loss of p53 , and of skin tumors in p53-deficient epidermis .
[159]
[158]
More recently, Levine et al. used a mouse model of intestinal neoplasia with a single truncated allele of
[160]
Min
the adenomatous polyposis coli (APC ) tumor suppressor and generated a doxycycline-inducible mouse
model exhibited increased levels of PLK4 (APC Min/+ ; Plk4 ), which resulted in centrosome amplification
Dox
Dox
and aneuploidy. Notably, the APC Min/+ ; Plk4 exhibited higher intestinal tumor incidence compared to the
APC but no greater tumor burden. Therefore, these results demonstrate that centrosome amplification has
Min
a role in tumor initiation but not in tumor progression. To investigate if centrosome amplification can drive
spontaneous tumorigenesis, Levine et al. also developed a ROSA26-rtTA; tetO-Plk4 mouse model that
[160]
expressed Plk4 in multiple mouse tissue upon doxycycline treatment. These mice developed lymphomas,