Why is cyclin D1 upregulated during cellular senescence?

The abstract below is from a recent publication of mine which shows my recent findings on the search for biomarkers of cellular senescence, which formed part of my PhD. It presents data which suggests that cyclin D1 overexpression can be used to detect senescent vascular smooth muscle cells and fibroblasts. However, there is little discussion on why cyclin D1, a known protein involved in cell cycle progression, is up-regulated at senescence (a state of growth arrest). The following is a literature review which may provide insight into why cyclin D1 is up-regulated in some senescent cell types.

Cyclin D1 Overexpression Permits the Reproducible Detection of Senescent Human Vascular Smooth Muscle Cells

The senescence of mitotic cells is hypothesized to play a causal role in organismal aging. Cultures of normal human cells become senescent in vitro as a r
esult of a continuous decline in the mitotic fraction from cell turnover. However, one potential barrier to the evaluation of the frequency and distribution of senescent cells in tissues is the absence of a panel of robust markers for the senescent state. In parallel with an analysis of the growth kinetics of human vascular smooth muscle cells, we have undertaken transcriptomic comparisons of early- and late-passage cultures of human vascular smooth muscle cells to identify potential markers that can distinguish between senescent and growth-competent cells. A wide range of genes are upregulated at senescence in human vascular smooth muscle cells. In particular, we have identified a 12-fold upregulation of expression in the cyclin D1 message, which is reflected in a concomitant upregulation at the protein level. Quantitative cytochemical analysis of senescent and growing vascular smooth muscle cells indicates that cyclin D1 reactivity is a considerably better marker of replicative senescence than senescence-associated β-galactosidase activity. We have applied this new marker (in combination with Ki67, COMET, and TUNEL staining) to the study of human vascular smooth muscle cells treated with resveratrol, a putative anti-aging molecule known to have significant effects on cell growth.

An understanding of why cyclin D1 is up-regulated at senescence may provide further insight into the molecular pathways governing cellular senescence (and cancer).

Interestingly, two variants of cyclin D1 exist (Solomon et al, 2003), these variants, designated cyclin D1a and cyclin D1b have been shown to differ in their behavior. Cyclin D1b does not possess the Thr286 phosphorylation site required for nuclear export and regulated degradation (Knudsen, 2006). As a result, the cyclin D1b protein appears to be constitutively localised in the nucleus, whereas cyclin D1a is exported to the cytoplasm during S-phase. Despite enhanced nuclear localisation, it was found that cyclin D1b is a poor regulator of RB phosphorylation/inactivation.

Probably the most interesting findings on cyclin D1 up-regulation come from Berardi et al (2003). This group identified a novel transcriptional regulatory element in the 5’-untranslational region of the cyclin D1 gene that differentially suppresses cyclin D1 expression in young versus senescent fibroblasts. Abundant protein complexes were found to be forming with young cell nuclear extracts compared with senescent cells nuclear extracts and binding was maintained in quiescent cells, showing that loss of activity was specific to senescent cells and not an effect of cell cycle arrest. These findings thus suggest that loss of transcriptional repressor activity may contribute to the up-regulation of cyclin D1 during cellular senescence.

Alt et al (2002) suggests that the accumulation of cyclin D1 at senescence may be due to elevated levels of p21. Evidence suggests that p21 promotes nuclear accumulation of cyclin D1 complexes via inhibition of cyclin D1 nuclear export. However, another study has demonstrated that oncogenic Ras promotes the accumulation of p21 by elevating the levels of cyclin D1 (Coleman et al, 2003). Colman and co-workers also found that this increase in cyclin D1 was sufficient to inhibit proteasome-mediated p21 degradation. Knock-down of cyclin D1 by RNA interference confirmed that RAS-induced p21 stabilisation was dependent upon cyclin D1 expression. They also showed that p21 directly binds to the C8α subunit of the 20S proteasome complex and that by competing for binding, cyclin D1 inhibits p21 degradation by purified 20S complexes in vitro. They therefore proposed that Ras stabilises p21 by promoting the formation of p21-cyclin D1 complexes that prevent p21 association with, and subsequently degradation by, the 20S proteasome. In some circumstances, the activation of Ras leads to cell cycle arrest similar to that observed with replicative senescence (Mason et al, 2004). It is therefore possible that at the onset of cellular senescence p21 is elevated, this in turn promotes nuclear accumulation of cyclin D1 which stabilises p21 and allowing it to accumulate further.

It was mentioned previously that the cyclin D1b variant lacks the Thr286 phosphorylation site required for nuclear export and degradation and is a poor catalyst for pRb phosphorylation. It is possible that during the replicative lifespan of a cell, this cyclin D1b variant could gradually accumulate within the cell nucleus binding and stabilising p21 until it reaches a threshold where all cyclin D1-Cdk complexes have bound to p21, triggering growth arrest. However, it has been reported that this cyclin D1 variant does not accumulate in cells and exhibits stability comparable with cyclin D1a (Soloman et al, 2003).

Senescent cells that express high levels of cyclin D1 are unable to phosphorylate pRb in response to mitogenic stimuli (Atadja et al, 1995). This study showed that the lack of pRb phosphorylation at senescence occurred when virtually all cyclin D1-Cdk complexes became associated with p21 (Dulic et al 1993). Therefore, it seems that low levels of cyclin D1 play a positive role in cell cycle progression by phosphorylating and thus neutralising the inhibitory activity of pRb. However, when cyclin D1 is elevated, it has a negative effect on the cell cycle by stabilising p21 and inhibiting cyclin D1-Cdk complexes from phosphorylating pRb, resulting in cell cycle arrest.

Increased p21 expression appears to only initiate telomere dependent senescence, but later, the senescent state is maintained by p16, at which point p21 is down-regulated (Stein et al, 1999). If p21 is down-regulated, this may also result in the down-regulation of cyclin D1.

If p21 expression is required to stabilise and consequently up-regulate cyclin D1, is cyclin D1 up-regulated in cells which are not dependent upon telomere shortening in order to senesce? Telomere independent senescence appears to trigger the up-regulation of p16 alone. This would suggest that cyclin D1 cannot be detected in these cell types. Interestingly, Opitz et al (2001) found that cyclin D1 overexpression alone was enough to extend the replicative lifespan of normal oral keratinocytes, a cell type known to senesce by telomere-independent mechanisms. Therefore, cyclin D1 overexpression in these cells have the opposite impact on cell state.

Typical dual staining of MRC-5 fibroblasts with cyclin D1 (FITC, green) and Ki67 (TRITC, red). DAPI (counterstain) is blue.

1 comment:

Sav said...

Thanks! This really helped!

The main focus of ageing research is to prevent/combat age-related disease and disability, allowing everyone to live healthier lives for longer.