Christina Maria Nelson^[1], Faculty of Science, Monash University

Abstract

The role of cell polarity in tumours which arise from epithelial cells is now well accepted. In the last decade, there has been increasing evidence to indicate that cell polarity is a critical regulator of leukaemia. Intrinsically linked to cell polarity and, more importantly, leukaemia stem cell (LSC) polarity is asymmetric cell division (ACD). In model organisms, such as Drosophila melanogaster, ACD has been shown to be an oncogenic-initiating event. Consequently, observations that ACD has a functional role in hematopoietic stem cell (HSC) development which is innately linked to LSC development could have profound implications on dissecting the role of cell polarity in leukaemogenesis. The article will discuss cell polarity as well as addressing ACD to review how dysregulation of these mechanisms might contribute to leukaemogenesis.

Keywords: Leukaemia, asymmetric cell division, cell polarity, self-renewal, hematopoietic stem cells, planar cell polarity pathway.

Introduction

Correct establishment of cell polarity - the asymmetric distribution of different cellular components - is essential for self-renewal and differentiation of stem cells (Lee and Vasioukhin, 2008: 1141). It is now recognised that disruption to cell polarity mechanisms plays an important role in the activation and proliferation of solid tumours due to studies on neuroblasts in the model organism Drosophila melanogaster (Caussinus and Gonzalez, 2005). However, the potential function of cell polarity in leukaemogenesis is less well explored. Currently, there is a growing body of research examining the link between cell polarity and leukaemogenesis; however, deciphering its role has not been straightforward. The article will discuss our current understanding of cell polarity and ACD in leukaemogenesis, and how disruption to these mechanisms might impact on different leukaemia lineages. First, the article will discuss the role of the Scribble polarity complex as a polarity module, and the planar cell polarity (PCP) pathway as illustrations for how loss of cell polarity might impact on leukaemogenesis.

Next, the article will discuss cell polarity in relation to asymmetric cell division (ACD) as a possible leukaemia-initiating event. The term ACD refers to the polarised distribution of fate determinants, components which determine the nature of a cell, into two daughter cells of a dividing cell (Zimdahl et al., 2014: 245). To review a possible functional role of ACD in leukaemogenesis, ACD and its key regulators will be examined in chronic myeloid leukaemia (CML) and acute myeloid leukaemia (AML). In addition, the role of ACD in cells of the immune system; namely, B and T cells (or lymphocytes) will also be addressed in respect to lymphoblastic leukaemia.

Finally, there is growing evidence to indicate that LSCs depend on similar niche signals to HSCs (Tabe and Konopleva, 2014: 767). To provide context, research to date has supported a functional role of ACD in HSC self-renewal with one of the two daughter cells always remaining an HSC (Riether et al., 2015: 188). Therefore, considering that HSCs maintain a self-renewing population via ACD then this might suggest that the proliferative state of LSCs are also maintained by ACD. The article will also discuss how LSCs might maintain a self-renewing population during leukaemia progression via ACD and how its key regulators might be important in leukaemogenesis. In summary, the article will address the role of cell polarity as well as a possible functional role for ACD in leukaemia.

Cell polarity and its role in leukaemia

The Scribble polarity complex

The Scribble polarity complex consisting of Scribble, Lethal giant larvae (Lgl) and Discs large (Dlg) maintains apicobasal polarity and regulates cell-fate determination (Zeitler et al., 2004: 1138). Members of the Scribble polarity complex were originally identified as tumour suppressors in epithelial tissues in the model organism Drosophila melanogaster (Royer and Lu, 2011). Although there is increasing evidence for a functional role of Scribble in leukaemogenesis there is still much work to be done to fully elucidate its role as a tumour suppressor. One such example is the member of the Scribble polarity complex, lethal giant larvae homolog 1 (Llgl1), which has been implicated as a critical cell polarity regulator of acute myeloid leukaemia (AML). Heidel et al. (2013) examined 83 patients below the age of 60 with cytogenetically normal AML (CN-AML) using gene expression analysis to analyse regulator gene defects. The analysis demonstrated that low expression of Llgl1 as measured by Affymetric gene arrays was associated with substantial decreased overall survival in normal AML where P=0.0092 (Figure 1). This suggests that decreased expression of Llgl1 in human AML cells increases the self-renewal capacity and fitness of HSCs which in turn is correlated to poor prognosis in AML patients.

Figure 1: Llgl1 repression leads to the development of human AML. Low expression of Llgl1 was associated with a greater decrease in the overall survival when compared to the high expression Llgl1 cohort in 83 of AML patients below the age of 60 years.

In contrast to the findings of Heidel et al. (2013), decreased Llgl1 expression did not contribute to the onset of mammalian B and T cell leukaemias driven by constitutive c-Myc, Notch or Jak2 signalling (Hawkins et al., 2014: 1-2). The study conducted by Hawkins et al. (2014) demonstrated that decreased Llgl1 expression did not alter HSC lineage development and conditional deletion of Llgl1 did not induce leukaemia in mice reconstituted with either NotchICN∆Ram∆P transduced or EμMyc/TEL-JAK2 foetal livers. This suggests that the role of Llgl1 might be restricted to a limited set of oncogenic lesions during development of hematopoietic malignancies. In addition Hawkins et al. (2014) demonstrated that Llgl1 deficiency alone is insufficient to induce leukaemia. This differs in the findings described by Heidel et al. (2013), which showed that decreased expression of Llgl1 correlates to poor prognosis in human AML patients. This indicates that further research is required to determine whether the development of different leukaemia lineages might be controlled by different members of the Scribble polarity complex.

Surprisingly, and in contrast to findings in epithelial tissues, there was little or no impact of deletion of Dlg1 on haematopoiesis (Humphries et al., 2012). This contrasts with a recent study suggesting that deletion of Dlg1 contributes to the development of paediatric B cell acute lymphoblastic leukaemia (B-ALL) (Sandoval et al., 2013: 433). In the study one important interaction examined was that between PTEN protein and a member of the Scribble polarity complex, Dlg1. The PTEN protein acts as a tumour suppressor and thus regulates cell division by ensuring that cells do not grow or divide too rapidly by opposing the effects of PI3K activation. The results showed that in early stage B-lineage cells the loss of Dlg1 did not alter the level of PTEN mRNAs but did result in a substantial decrease in the level of PTEN protein leading to excessive PI3K signalling and proliferation of large pre-B cells (Figures 2a and 2b). These observations suggest a possible role for Dlg1 to stabilise PTEN and thus to address this, pulse-chase experiments were carried out with results indicating that the rate of degradation of PTEN in Dlg1 deficient cells was markedly faster compared to control cells (Figure 2c). This instability is thought to occur as a result of the formation of a truncated protein lacking the C-terminal PDZ binding domain which decreases the stability of PTEN. Overall it was shown that Dlg1 loss disrupts PTEN causing increased proliferation of B-ALL. Therefore presumably the loss of Dlg1 can be used as a potential model for leukaemia suppression.

Figure 2: Dlg1 is required for stabilisation of PTEN. (a) PTEN protein levels in day 4 pre-B cell IL-7 cultures via Western blotting. (b) PTEN levels similar between WT and KO. (c) Pulse-chase experiments in which cells were pulsed with [35S]methionine for 30 minutes, and then chased for 3 and 6 hours. Results indicate degradation of PTEN is greater in KO comparted to WT.

In a pivotal study conducted by Hawkins et al. (2016) it was shown that the tumour regulatory functions of Scribble are context dependent, and, furthermore, the data provided evidence that overexpression of a mutant form of Scribble can act as an oncogene in Eμ-myc-driven lymphoma. This was achieved by using the Eμ-myc model of Burkitt's lymphoma which simulates constitutive activation of the oncogene c-myc. Myc (c-myc) is a regulator gene encoding a transcription factor which, when expressed, plays a role in cell cycle, differentiation and apoptosis. The results indicated that loss of Scribble does not function as a tumour suppressor in Eμ-myc-driven lymphoma but instead the onset of the disease was delayed and it was shown that fate determination driven by Scribble is not an essential function of lymphomagenesis. These results were surprising and suggests that Scribble does not function as a tumour suppressor as originally understood. Collectively these studies which have examined members of the Scribble polarity complex indicate that members might function differently depending on the hematopoietic lineage.

PCP pathway implicated in leukaemia

One such signalling pathway proposed to contribute to leukaemogenesis is the PCP pathway, a conserved pathway regulating cell polarity and migration. In a study by Kaucka et al. (2013), the key PCP pathway components were upregulated in B lymphocytes of patients with chronic lymphocytic leukaemia (CLL), suggesting that the PCP pathway is a critical regulator of CLL. Analysis of key PCP protein levels during progression of CLL by Western blotting confirmed increased PCP protein expression including Vangl2, Dvl2, Dvl3, Prickle1, and CK1ε (Figures 3a and 3b). In addition, when samples of six patients with CLLs collected at two different time points during CLL progression were compared, the overall expression of PCP proteins had increased across the two time points. Following this, it was also demonstrated that a PCP ligand, Wnt5a, in synergy with CK1, which phosphorylates a key PCP protein Dv1, increased cell polarisation and directional migration of primary CLL cells across the PCP pathway. These results indicate that dysregulation of PCP pathway can serve as a navigator of invading CLL cells which in turn contributes poor prognosis in CLL patients.

Figure 3: PCP genes and proteins are upregulated in CLL cells. (a) The amounts of the indicated PCP proteins analysed by Western blotting in B cells of patients with CLL and healthy controls. Patients with CLL have greater expression of PCP pathway proteins. (b) Quantification of data in (a).

The role of asymmetric cell division in leukaemia

Lis1 as a critical regulator of ACD

Although ACD is critically important in generating diversity during development, its dysregulation can also promote oncogenesis (Bajaj et al., 2015: 1). Not surprisingly, as ACD is intrinsically linked to cell polarity, critical regulators of ACD are the polarity regulators. One such cell polarity regulator of ACD is Lis1, a dynein-binding protein co-ordinating mitotic spindle attachment during cell division (Siller and Does, 2008: 1). However, the functional role of Lis1 as a critical regulator of ACD in HSC fate and leukaemogenesis is relatively unexplored. Recently, Zimdahl et al. (2014) reported that Lis1 is an integral protein in HSC development and demonstrated its directed control of ACD in blast-crisis myelogenous leukaemia (blast-crisis CML) and therapy-resistant de novo AML. Generation of blast-crisis CML was achieved by co-infecting LSK cells with viral constraints from BCR-ABL and NUP98-HOXA9 and transplanting them into sublethally irradiated recipients. Results indicated that all mice transplanted with control leukaemia-propagating cells developed leukaemia; however none of the mice transplanted with cells that conditionally lost Lis1 developed leukaemia (Zimdahl et al., 2014: 249-50). This showed that Lis1 is a critical regulator in the establishment of CML.

Increased differentiation can be a consequence of defects in ACD and thus it was shown that loss of Lis1 contributed to the altered polarisation of fate determinants (Zimdahl et al., 2014: 248). To establish ACD, Numb distribution relative to spindle orientation was used. A similar impairment also occurred in AML mouse models as all mice transplanted with a control died of AML within 3 weeks; however, only approximately 40% of mice transplanted with cells that conditionally lost Lis1 developed AML. At a cellular level, Lis1 deletion resulted in a fivefold increase in the number of differentiated leukaemic cells (Zimdahl et al., 2014: 250). To determine whether Lis1 is also required for human myeloid leukaemia, LIS1 was deleted in human samples of aggressive leukaemias. Deletion of LIS1 showed increased differentiation in the short term but impeded growth in the long term. These findings provide genetic evidence that the balance of ACD and its regulators are critical for human leukaemia growth and propagation.

Musashi-2 and its role in ACD

Musashi (Msi) was initially examined as a critical regulator of ACD during sensory neuron development in the model organism Drosophila melanogaster (Nakamura et al., 1994: 67). Msi is part of a family of binding proteins which, among other things, control the regulation of symmetric cell division (SCD) and ACD. In particular, Msi-2 has recently been associated with ACD regulation in normal HSCs and in myeloid leukaemias. In a study conducted by Kharas et al. (2010), gene expression profiling was used to examine the effect of MSI-2 on 2 ᵡ AML and 2 ᵡ CML-BC cell lines. It was proposed that loss of MSI-2 expression correlated with increased survival. Similar to the role of Lis1 in ACD regulation as described by Zimdahl et al., it was found that loss of Msi-2 could be an AML initiating event. This data provides further support for the comparatively new aspect of ACD in leukaemogenesis.

Msi-2 expression also influences the prognosis of patients suffering from lymphoblastic leukaemias via the dysregulation of ACD in the LSC niche. In a study conducted by Mu et al. (2013), Msi-2 expression of 116 adult B-ALL patients was measured by real time polymerase chain reaction (PCR). Kaplan-Meier analysis was then used to measure the fraction of subjects who live for a certain amount of time after treatment. The analysis showed that patients with high Msi-2 expression had inferior overall survival (P=0.004) and relapse-free survival (P=0.018) compared to patients with low Msi-2 expression. However, similar to the role of Llgl1, the observed Msi-2 defect was also more severe in the myeloid lineage compared with the lymphoid lineage (Park et al., 2014: 75). In other words, loss of Msi2 affects the engraftment of myeloid and lymphoid lineages but the loss of Msi2 has a greater impact on the myeloid lineage. This again suggests, as previously discussed, that different polarity regulators may dictate the fate of different leukaemia lineages. This indicates that further research needs to be conducted to determine whether dysregulation of ACD is a leukaemia-initiating event.

Role of ACD in cells of the immune system

The requirement for expansion of many individual T cell clones opposed to expansion of the entire T cell population during mammalian T cell development indicates a possible role for ACD (Pham et al., 2015: 933). Since leukaemic cells (LSCs) express different antigens that are able to activate CD4+ and CD8+ T cells (Riether et al., 2015: 187), then presumably activation of these T cells can regulate the expansion of the LSC pool during leukaemogenesis. Therefore, elucidating the role that ACD plays in both T cell development and activation is important to gain a complete understanding of its role in leukaemia. One of the first studies implicating ACD as an important regulator in the generation of effector and memory T cells was reported by Chang et al. (2007). Later Pham et al. (2015) showed that ACD occurs specifically during the β-selection checkpoint of T cell development at the DN3A stage to regulate fate decisions. This is of significance as the β-selection checkpoint has been shown to be of importance in the prevention of leukaemia (Yui and Rotheberg, 2014: 530). In addition it has been shown that Scribble is unevenly distributed during the generation of effector and memory T cells via ACD (Oliaro et al., 2010: 668). This provides further evidence that ACD is critically regulated by cell polarity proteins which in turn are asymmetrically distributed in T cells during cell division (Figure 4). Together these findings indicate that ACD has a role in the development of T cells and thus potentially dysregulation of ACD can lead to the development of leukaemias arising from T cells.

Figure 4: The asymmetric localisation of proteins during asymmetric cell division.

Hawkins et al. (2013) undertook a genetic and functional analysis of the relationship between the Scribble complex as a critical regulator of B cell activation and proliferation via ACD. To elucidate whether there was a link between ACD and B cell activation, Hawkins et al. (2013) established a polarised source of CD40, a protein found on antigen presenting cells (B cells), stimulation by coating beads with αCD40 antibody. The results showed that B cells which underwent multiple divisions by ACD were independent of Scribble (Figure 5b) and that there were no differences in division on days three and four (Figure 5a). Surprisingly, Hawkins et al. (2013) found that loss of Scribble had no effect on B cell differentiation and conclude that ACD is not required for B cell activation and differentiation. This contrasts to T cells in which it was found that ACD has a role in mammalian T cell development. The result of Hawkins et al. (2013) provides no support for ACD in B lymphoblastic leukaemias and thus further research is required to determine the role of ACD in immune cells.

Figure 5: Scribble B cells respond to polarised stimulation and display normal migration. (a) Quantification of cell numbers per division on days 3 and 4 with Scribble (Scribble+/+) and without Scribble (Scribblenull/null). (b) Quantified with relation to cell number per division showing that differentiation of B cells is not independent on the expression of Scribble.

ACD and self-renewal of hematopoietic stem and progenitor cells

Self-renewal is linked to stem cell division and despite models suggesting that these processes involve ACD there is still little understanding of the molecular details. To address whether ACD is a critical regulator of HSC self-renewal, altered HSC gene expression data was examined by established in vitro and in vivo screen, revealing six regulators of HSC fate while also exhibiting ACD at cell division: Ap2a2, Gpsm2, Tmod1, Kif3a, Racgap1, and Ccnb1 (Ting et al., 2012). In particular, live-cell video microscopy of the endocytic protein Ap2a2 showed asymmetric segregation during HSC and progenitor cell cytokinesis. This serves as a basis for further studies to determine whether ACD, HSC fate and self-renewal are systematically linked. In summary, the study provides further evidence for a functional role of ACD in HSC maintenance (self-renewal) which might contribute to the control of the HSC pool.

Ramifications of increased rates of ACD

There is increasing evidence supporting the theory that LSCs adopt similar niche signals to HSCs. This means that the LSC population might be maintained by adopting ACD as a regulatory mechanism to maintain a self-renewing LSC pool during leukaemogenesis. Research to date does not prove an effect of Msi-2 on self-renewal capacity (de Angres-Aguayo et al., 2011). Given Msi-2 has a functional tole as a regulator of ACD, it is possible that dysregulation of ACD due to Msi-2 could influence LSCs self-renewal capacity. In addition two recent studies showed that Msi2 cooperates with fusion oncoproteins in the induction of acute myeloid leukaemia by inhibiting the ACD determinant Numb (Kharas et al., 2010; Ito et al., 2010). This suggests that increased rates of ACD due to the loss of Msi-2 could result in disruption to the division of fate determinants such as Numb and therefore potentially contribute to the maintenance of the LSC pool.

There is increasing evidence that the cell-fate determinant Numb plays a functional role as a tumour suppressor in distinct types of human cancers (Gomez-Lopez et al., 2014: 575). Reduced Numb levels are also observed during the progression of CML to blast-crisis which is associated with increased Notch signalling and reduced TP53 activity (Ito et al., 2010: 766). In addition Zimdahl et al. (2014) showed that decreased Lis1 expression contributes to a complete reversal in the pattern of inheritance, with twofold more cells undergoing ACD. The increased level of ACD with more cells inheriting high levels of Numb generates a greater number of differentiated cells with each division leading to exhaustion of the stem cell pool. Furthermore, as mentioned previously, loss of Llgl1, a critical regulator of ACD also leads to expansion of HSCs and enhanced repopulation capacity after serial transplantation (Heidel et al., 2013: 16). Consequently, as leukaemia is characterised by increased self-renewal capacity then this might explain a possible functional role for ACD in leukaemogenesis.

Conclusion

This article examined the Scribble polarity complex and the PCP pathway to provide evidence of a possible role for cell polarity in leukaemogenesis. Current research on members of the Scribble polarity complex, particularly Llgl1, has been controversial and therefore further research is required to fully dissect the role of the Scribble polarity complex in different leukaemia lineages. However, the role of the PCP pathway as a cell polarity regulator does provide definitive evidence as to its role in lymphoblastic leukaemia. To determine whether ACD might be a leukaemia-initiating event the article examined its key regulators. It was shown that dysregulation of ACD regulators via decreased expression of Lis1 and Msi-2 provides evidence for ACDs role during myeloid and lymphoblastic leukaemia. It is also evident that one of the most convincing demonstrations for a functional role for ACD is in T lymphocytes which activate LSCs during their proliferative stage. In summary, a definitive explanation for the role of cell polarity and ACD in leukaemogenesis requires a complete genetic map of the effect on disruption of important cell polarity regulators and ACD.

---

Acknowledgments

I would like to thank Melissa Honeydew for her advice during the completion of my literature review as part of the scientific practice and communication (advanced) course in first semester of 2015.

List of illustrations

Figure 1: Llgl1 repression leads to the development of human AML. Image reproduced with permission from F. Heidel et al. (2013) regarding Llgl1 expression and human AML.

Figure 2: Dlg1 is required for stabilisation of PTEN. Image reproduced with permission from G. J. Sandoval et al. (2013) regarding PTEN levels.

Figure 3: PCP genes and proteins are upregulated in CLL cells. Image reproduced with permission from M. Kaucka et al. (2013) regarding planar cell polarity genes in CLL cells.

Figure 4: The asymmetric localisation of proteins during asymmetric cell division. Image reproduced with permission from S. Russell (2008)regarding asymmetric cell division.

Figure 5: Scribble B cells respond to polarised stimulation and display normal migration. Image reproduced with permission from E. Hawkins et al. (2013) article regarding B lymphocytes and asymmetric cell division.

Notes

^[1]Christina Nelson studies a Bachelor of Science at Monash University completing her third year doing a major in Biochemistry and a minor in Chemistry.

References

 

 

To cite this paper please use the following details: Nelson, C.M. (2016), 'The Role of Cell Polarity in Leukaemogenesis', Reinvention: an International Journal of Undergraduate Research, Volume 9, Issue 1, http://www.warwick.ac.uk/reinventionjournal/archive/volume9issue1/nelson Date accessed [insert date]. If you cite this article or use it in any teaching or other related activities please let us know by e-mailing us at Reinventionjournal at warwick dot ac dot uk.