MBPhD Programme




 July 2013

Current Position (2018/19):

  • Post PhD: MBBS Year 6 - Preparation for Practice

PhD title:

Quantitative biology of cell cycle decision making

Principal Supervisor:

Sir Paul Nurse, CRUK LRI

Funding Source:

Cancer Research UK, LRI

Description of Project:

In this thesis, I developed multiple new single cell methodologies for the interrogation of cell cycle decision making in S. pombe. Using these tools, I found new principles relevant to how cells measure their size and decide to move through different cell cycle transitions. Additionally, I used these new tools to investigate the functional relevance of APC/C mutations found in cancer.

New tools and methodologies

I developed a new set of image analysis algorithms for the single cell segmentation and tracking of S. pombe from brightfield images. The algorithms are relatively independent of cell morphology, so can be used for the quantification of cell morphology in various mutant strains. These will hopefully be of use to the S. pombe community.

In this thesis, I present a set of fluorescent proteins, and the computational algorithms necessary to quantify them to automatically assign cell cycle phase to images of single cells. This approach can identify the relationship between cell size and cell cycle decision making in different genetic backgrounds. Other S. pombe researchers who want to quantify the effect of a gene deletion or mutation on cell cycle decision making could employ these tools.

A new single cell CDK activity reporter, and the algorithms required to quantify its levels are presented in this thesis. I have also developed two new experimental protocols for the measurement of CDK activity on entry to mitosis — one using a time-lapse block and release assay, the other using imaging flow cytometry. These methods could be of use to other researchers who want to quantify the effect of their mutation or gene deletion of interest on CDK activation in mitosis.

I developed a set of tetracyline inducible promoters for fission yeast. Current inducible promoters in S. pombe either affect cell physiology, have low dynamic range or have switch like dose response kinetics. Tetracycline does not affect S. pombe cell growth, the new promoters are activated in minutes and feature linear dose response kinetics. These promoters add to the S. pombe genetic toolkit.

New biological insights

Using the tools described above, I have implicated the levels of cyclin B in controlling S. pombe cells size. The model of cell size presented could have implications for other organisms too.

I have shown that some cell size dependent pathway directly influences CDK activity outside any known canonical regulators. This result demonstrates that cell size can directly impact CDK activation.

S. pombe is quite resistant to CDK1 inhibition, and I show a potential molecular mechanism for this robustness. Understanding how cells survive could be translationally important, due to lacklustre performance of CDK1 inhibitors in the clinic.

Finally, I modelled APC/C mutations found in cancer in S. pombe. I showed that some of these mutations impacted APC/C activity. Cancer genome sequencing has generated a wealth of data on the mutations found with malignancies, although it is difficult to identify ones of functional relevance. Mutations in well-conserved genes could potentially be screened in S. pombe for their importance in humans.

Why are males and females different?

Men are four times more likely than women to develop autism spectrum disorders (Robinson et al. 2013; Geier et al. 2012). Conversely, women are more susceptible than men to many autoimmune diseases, including rheumatoid arthritis and Graves' disease (Rubtsova et al. 2015). Such differences have historically been attributed to the action of sex steroid hormones, primarily expressed from the gonads and causing sexual differentiation throughout the body, e.g. (MacLusky & Naftolin 1981; Wilson et al. 1981). However, it is becoming increasingly recognised that X and Y genes are differentially expressed in all somatic tissues, thus directly contributing to phenotypic differences at the cellular, tissue and organism level (Cortez et al. 2014; Bellott et al. 2014).

The mammalian sex chromosomes evolved from a pair of autosomes around 180-200 million years ago aegalian & Page 1998). Subsequently, the X chromosome has largely maintained its gene content, whereas the Y chromosome has become relatively gene poor (Rozen et al. 2003; Cortez et al. 2014; Bellott et al. 2014). However, a subset of genes have maintained both X and Y linked copies, the so-called X-Y gene pairs (Bellott et al. 2014; Cortez et al. 2014). Notably, many of these X-Y pair genes show ubiquitous expression, and are involved in key regulatory functions, such as transcription initiation, splicing, and translation (Bellott et al. 2014). The necessity of these functions likely explains the X-Y pair genes' longevity across evolutionary time, and a relative sensitivity to dosage. It is therefore entirely feasible that such genes, widely expressed as sex specific protein isoforms in both males and females, could underlie at least part of the commonly recognised sexual dimorphism in health and disease.

Using the mouse as a model system, we hypothesised that either two X-linked copies (in females), or one X-linked and one Y-linked copy, of each of these genes would be required for a normal phenotype. Furthermore, we hypothesised that at least one copy would be necessary for postnatal survival. We were interested to test the postnatal survival hypothesis, and begin to investigate the functions of these X-Y gene pairs. Such information will not only further our understanding of disease pathogenesis, but would also be highly valuable in future therapeutic design and screening.

By setting out to create novel mutant alleles for four of the X-linked genes from the X-Y pairs — Kdm5c, Kdm6a, Ddx3x, and Eff 2s3x — using CRISPR-Cas genome editing, we elucidated a number of significant findings. Kdm5c and its Y-linked homologue Kdm5d are functionally interchangeable — one or the other is required for postnatal survival in mouse. In contrast, our evidence suggests that one copy each of the X-linked genes Kdm6a, Dekdx, and E#2s3A- is required for postnatal survival. The Y-linked homologues, Uo, Debc3y, and Eigs3y, are no longer able to compensate for the loss of their X-linked evolutionary ancestor. These data suggest a significant degree of functional divergence in the mouse X-Y gene pairs that are highly conserved at the amino acid level. We conclude that, whilst at the sequence level these genes are conserved, the regulatory regions have likely evolved significant differences between the chromosomes, and should be the focus of further investigation into male-female differences.

Sex chromosome aneuploidy in the placenta

Females with a single X chromosome, and no second sex chromosome (XO), have Turner syndrome. This aneuploidy has an incidence of approximately 1:2000 live births (Hook & Warburton 2014). Whilst the clinical phenotype is highly variable from woman to woman, almost all are born small, and rarely attain an adult height of more than 150cm if not treated with growth hormone. Strikingly, it has been estimated that 99% of embryos conceived that have an XO karyotype are spontaneously aborted during pregnancy (Hook & Warburton 1983). This effect has been attributed to an absolute requirement for female embryos to have two X chromosomes at a vital, unknown stage in development; and the 1% surviving are hypothesised to be mosaic, i.e. some cells have two sex chromosomes and rescue the embryo (Hook & Warburton 1983). Many studies have been carried out to identify these so-called cryptic mosaic patients, and a small number have been found, but far from all surviving patients with a Turner syndrome diagnosis are overtly mosaic (Hook & Warburton 2014). This has led to the suggestion that the mosaic tissue may be the shortest lived of organs, the placenta (Urbach & Benvenisty 2009). A number of candidate genes have been identified in human cell lines, though supporting evidence is currently lacking.

Interestingly, like humans, female mice with an X0 karyotype are also small compared to normal female littermates, both during embryonic development and early postnatal life (Thornhill & Burgoyne 1993). These mice can therefore be used as a model for investigating the effects of aneuploidy on embryonic development. We derived stem canines from early mouse placentas with either XPO and XMO karyotypes, i.e. carrying either a paternally inherited or maternally inherited X chromosome, in addition to XX control lines. We then performed RNA sequencing analysis on these lines, aiming not only to identify differences between euploid and aneuploid placental stem cells, but also to look for differences in gene expression dependent on the parental origin of the X chromosome, i.e. imprinted genes. We were able to highlight a small number of novel candidates for imprinted genes that are currently undergoing validation experiments. Most interestingly, we observed >5000 genes differentially expressed between XX and both XPO and XmO placenta cell lines. We believe that such global transcriptional dysregulation in the aneuploid state may at least partly underlie the phenotype observed in the mouse and, if present and validated in human, may fundamentally change our currently embryo-centric view of Turner syndrome.