research reports

University of York

2005/6


Regulation of initiation of mammalian DNA replication by Ciz1
Dr Dawn Coverley, Dr Faisal Abdul Rahman, Dr Justin Ainscough,
Ms Heather Sercombe, Mrs Irene Watson and Mr Bjorn Gerlach

Summary
We have built on last year's work which showed that Ciz1 plays a positive role in DNA replication in two ways. First, we have dissected the domains structure of Ciz1 and shown that it interacts with the nuclear matrix through C-terminal domains that are distinct from replication-activity encoding domains. Second we have continued to analyse the complex issue of Ciz1 splice variant expression and their regulation during development and disease. We have focused on a polypyrimidine tract expansion that recurs in early passage cell lines from Ewings Sarcoma Family tumours and shown that it drives exon skipping events. Exclusion of this exon from Ciz1 reporter constructs alters the sub-cellular distribution of Ciz1 protein, while retaining activity in cell-free assays, raising the possibility that Ciz1 mis-splicing may influences the spatial organisation of initiation in mammalian cells.

Introduction
We identified Ciz1 based on its behaviour in cell-free DNA replication experiments (Coverley et al 2002) and its antigenic similarity to the DNA replication protein Cdc6 (Coverley et al 2005). Ciz1 stimulates DNA replication in cell-free and cell-based assays and is present throughout the cell cycle as sub-nuclear speckles that are coincident with sites of DNA replication in S phase cells.

Aims
The aims of this project are 1) To understand the mechanism by which Ciz1 influences mammalian DNA replication, 2) To compare the behaviour and function of Ciz1 splice variants, 3) To describe Ciz1 splice variant usage in normal and cancer cells and evaluate the relationship between Ciz1 mis-splicing and cancer, and 4) To determine whether Ciz1 variants can be used as a target for selective cell killing.

Research progress

1. The DNA replication protein Ciz1 is anchored to non-chromatin nuclear structures via C-terminal sequences (manuscript abstract - submitted)

Ciz1 co-localizes with replication factories in S phase nuclei and is required for cells to enter S phase. Here, we show that a significant proportion of the Ciz1 resists salt and nuclease extraction, and that the resistant fraction is retained in nuclear foci. This suggests that Ciz1 is immobilized by interaction with non-chromatin nuclear structures, consistent with the so-called nuclear matrix. Analysis of GFP-tagged Ciz1 fragments demonstrates that nuclear immobilization is mediated by sequences in the C-terminal third of Ciz1, encoded within amino-acids 708 to 830. Immobilization via C-terminal interactions occurs in a cell cycle dependent manner, possibly during the G1-S phase transition to coincide with the reported point of action of Ciz1. Combined with previous work, which showed that Ciz1 stimulates DNA replication via sequences in its N-terminal half, the data suggest that Ciz1 is composed of two functionally distinct domains, and that C-terminal interactions localize its DNA replication function to specific sites in the nucleus. Thus, Ciz1 may contribute to the spatial organisation of DNA replication in mammalian cells.

Fig. 1 Endogenous Ciz1 (red) resists detergent, nuclease and high-salt extraction and is retained in nuclear foci via C-terminal interaction with the nuclear matrix. Nuclear DNA is blue. Method described in Ainscough et al

Fig. 2 Schematic representation of Ciz1 attached to an insoluble structure via a C-terminal 'anchor' (yellow circle). Red triangles indicate alternatively spliced sequences that are conditionally absent from the N-terminal (replication active) half of Ciz1. Speculative model showing possible role for Ciz1 as a mediator between matrix-associated replication factories (grey circles) and chromatin-associated replication proteins (grey barrels).

2. Regulated splicing of exon 4 influences the sub-nuclear distribution of the DNA replication protein Ciz1, and is disrupted by intronic mutation in Ewings Sarcoma Family Tumour cell lines (manuscript in preparation)

Ciz1 is alternatively spliced in several paediatric cancers, and some adult cancers (often with atypical exon boundaries). Where possible we are testing whether these distinctive splicing patterns represent the cell types from which the cancers were derived, and also looking at whether aberrant Ciz1 splicing could alter Ciz1 replication activity and thus potentially influence cell proliferation.

So far we have focussed on Ciz1 exon 4 which is skipped in ESTs from several paediatric cancer-derived libraries. Our analysis has shown that Ciz1 transcripts lacking exon 4 are common to a panel of early passage Ewings Sarcoma Family tumour-derived cell lines, but not cell lines from the related paediatric cancer Neuroblastoma, or a range of other 'normal' cell lines.

Furthermore, we have identified an intronic polypyrimidine tract expansion in the Ewings Sarcoma cell lines, a phenomenon known to affect splicing in other disease genes including the Cystic Fibrosis Conductance Regulator gene intron 8 (Rave-Harel et al 1997), and MRE11 intron 4 (Ottini et al 2004). We have also used exon trap analysis to demonstrate that the expansion drives exon 4 skipping. Therefore, in this cancer Ciz1 mis-splicing is driven by mutation. Because our evidence indicates the presence of multiple polyT alleles in each Ewings sarcoma line, we have also used real-time PCR to look for evidence of Ciz1 copy number variation. Preliminary evidence suggests that this is the case in around half of the lines tested.

Functionally, Ciz1 exon 4- protein retains DNA replication activity in our assays but it fails to localise correctly in the nucleus, forming a diffuse rather than speckled pattern. Therefore, we propose that cells expressing Ciz1exon 4- may suffer spatially mis-regulated DNA replication. For the remaining duration of this project we will continue to focus on exon 4, and also ask whether Ciz1 mis-expression can impose sufficient replicative stress to activate the DNA damage response (see below)

Fig3 NIH 3T3 cells transfected with GFP-tagged mouse embryonic Ciz1 variant (ECiz1), with and without alternatively spliced exon 4. Both forms are active in DNA replication initiation experiments (not shown).

3. Related avenues

The cells response to excess Ciz1 Overexpression of ectopic Ciz1 initially stimulates initiation then inhibits nucleotide incorporation. This appears to be a transient response with some, but not all Ciz1 splice variants as flow cytometry shows that transfected cells do eventually recover and progress beyond S phase. H2AX phosphorylation increases in transfected populations suggesting that overexpressed Ciz1 may induce the DNA-damage response. We continue to investigate this avenue and hope to be able to devote more resources in the future.

Splice variant expression in lung cancers Related to the YCR-funded project above, we have also begun a commercially-funded project to study a second Ciz1 splice variant found predominantly in lung cancers. This has received funds from Yorkshire Forward through the Bioscience Yorkshire Enterprise Fellowship scheme, York University Research and Industry office and will be developed by University of York spin-out Cizzle Biotechnology Limited with funds from the White Rose Seedcorn Fund. YCR post-doctoral research fellow Dr Faisal Abdul Rahman has contributed to this work, which is being done primarily by myself and Mrs Irene Watson.

Antibodies We have embarked on production of two new anti-Ciz1 polyclonal antibodies, to complement the anti-N terminal fragment reagents already in the laboratory. Existing reagents fail to recognise exon 4 minus Ciz1, suggesting that this epitope may normally be exposed on the surface of the protein. Therefore, we are engaged in raising a new human/mouse anti-peptide reagent against the junction sequence generated when exon 4 is spliced out, as well as a new human/mouse anti-peptide reagent against C-terminal epitopes. The former reagent is expected to be used to study exon 4 minus expression in Ewings Sarcoma cells and mouse embryos, and the latter for sub-nuclear immuno-localisation studies.


YCR Cancer Research Unit
Director: Professor N.J. Maitland

Retargeting of baculoviral gene therapy vectors to infect the prostate

Mr V. Dussupt, Ms K. Tiemann, Ms S. Swift and Prof. N.J. Maitland

Collaborators:
Members of the EU-funded GIANT Program (co-ordinated by York)
Dr L. Stanbridge, Ms L. Hudson, Ms R. Nugent (YCR Cancer Research Unit, University of York)
Prof. C. Bangma, Dr R. Kraaij, Dr E. Schenk-Braat (Erasmus University, The Netherlands)
Dr A. Antson (York Structural Biology Laboratory, University of York)
Dr R. Mistry, Mr H. Amin, Dr A. Raymond (Pro-Cure Therapeutics, York)
+10 other collaborating laboratories

Baculoviruses are unique rod-shaped, membrane-bound viruses whose natural hosts are insects. However, we have previously shown that the virus particles can enter human cells, but are unable to replicate. To develop baculoviruses as gene therapy vectors it is necessary to re-target them to infect specific human cell types, including prostate cancers.

Targeting attachment to the cell membrane is one way to restrict expression to the prostate. Two short peptide ligands for receptors that are overexpressed in prostate cancer, i.e. gonadotrophin-releasing hormone receptor (GnRH) and the neurotensin receptor family, were inserted into the baculoviral envelope protein gp64 to achieve targeting. Virus attachment was also engineered via specific antibodies against surface proteins over-expressed in prostate cancer, by insertion of the protein A Z34C IgG binding domain into two positions in the gp64 protein sequence. Both insertions were capable of binding tightly to the Fc portions of monoclonal antibodies.

Large quantities of recombinant gp64 have also been generated, to facilitate X-ray crystallography and rational, 3D structure-based design of attachment peptide insertions.

To ensure specific therapeutic gene expression, transcriptional control has also been engineered, using native promoters modified to become androgen independent, or a synthetic truncated androgen receptor molecule (ARf), which activates prostate¡Vspecific promoters in the absence of androgens.

After testing in vitro, purified stocks of GnRH virus have now been injected into an orthotopic, xenografted human prostate tumour. The first generation humanised vectors showed widespread expression of a marker gene throughout the tumour mass.

Structure and functional studies of Human Papilllomavirus E2 proteins

Dr J.E. Burns, Ms E. Hernandez-Ramon, Ms H. Walker and Prof. N.J. Maitland

Collaborators:
Dr A. Antson, Prof K. Wilson (York Structural Biology Laboratory, University of York)
Dr I. Morgan (Institute of Comparative Medicine, University of Glasgow)
Dr S.V. Graham (Division of Virology, University of Glasgow.)
Dr S. Allen, Dr W. Zhang (School of Pharmaceutical Sciences, University of Nottingham)

HPV E2 proteins are the major virally-encoded regulators of viral transcription and replication. They are conserved 45 kDa nucleophosphoproteins which share a tripartite structure composed of an N-terminal module of around 200 amino acids (required for replication and transactivation functions) linked by a region of variable length and sequence to a C-terminal DNA-binding and dimerisation domain of around 80 ¡V 90 amino acids.

E2 represses transcription through the HPV p97 promoter in virally-infected cells. In persistently infected cells, the HPV genome may become integrated into host cell chromatin, with breakage usually occurring in the E1/E2 ORFs leading to unregulated expression of the E6 and E7 oncogenes and further progression towards a malignant phenotype.

Expression of E2 at detectable levels in human cervical tumour cells (where the E2 gene is disrupted) normally results in cell death. To study E2 function, we set out to make non-malignant keratinocyte cell lines stably transfected with an E2 expression plasmid, to express moderate levels of E2 under SV40 promoter control. As expected, most cells died rapidly, however, a number of clones that survived selection showed morphologies characteristic of differentiated keratinocytes. Transactivation assays showed that the cells did express E2 at low levels. Levels of expression of differentiation markers such as involucrin were similar to HaCaT cells that had been induced to differentiate in calcium-containing medium. We suggest that another function of E2 in the viral life-cycle may be to promote differentiation and hence stimulate virus production, which is dependent on keratinocyte maturation.

Identification of cancer stem cells in human prostate tumours

Dr A.T. Collins, Dr S. Bryce, Mr P. Berry, Mrs K. Hyde, Ms C. Brinham and Prof. N.J. Maitland

Collaborators:
Mr M. Stower (York Hospital NHS Trust)
Prof. F. Hamdy, Prof. N. Brown, Dr K. Linton (University of Sheffield)
Dr D. Hudson, Prof. C. Cooper (Institute of Cancer Research, Sutton, Surrey)
Prof. J. Southgate, Dr V. Knauper (Department of Biology, University of York)
Mr N.W. Clarke, Dr M. Brown (CRC Paterson Institute for Cancer Research and Christie Hospital NHS Trust, Manchester)
Dr G. Van Leenders, Dr G. Jenster (Erasmus University Rotterdam)
Dr G.van der Pluijm (Leiden University Medical Center)
Dr M. Cechini - bern
Dr J Schalken - runmc
Prof. D. Edwards (University of East Anglia)
Dr M. Campbell (University of Birmingham)
Prof. H. Leung (University of Newcastle)

Most cancer therapies treat the bulk of the tumour population, yet tumours (and particularly prostate cancers) are heterogeneous, composed of cells with different phenotypic characteristics and different proliferative and malignant potentials. Recent reports of cancer stem cells have prompted questions regarding the involvement of normal stem/progenitor cells in prostate tumour biology.

Stem cells are functionally defined as self-renewing and multipotent, generating the mature cell types of a particular tissue through differentiation. In malignancies, such as leukaemia, breast, brain and lung cancers, rare cells have been isolated with a remarkable potential for self-renewal, and these cells were distinct from the bulk of the tumour in driving tumour maintenance and growth.

Our group was the first to identify and isolate stem cells from non-malignant prostate epithelia, and by exploiting shared antigen expression between normal and cancer stem cells we have now been able to identify and isolate a population of cells with a phenotype corresponding to a cancer stem cell for prostate; showing evidence of self-renewal, proliferation, differentiation and invasion. These cells form a small subset of the population even after long-term culture and have a distinctive gene expression profile. We predict that both tumour heterogeneity and the overall androgen-sensitive phenotype of human prostate cancers are the consequence of differentiation from an AR-negative stem cell population. Furthermore, the phenotype suggests that the cancer stem cell is the source of residual disease after conventional therapies for prostate cancer. Whilst on-going in vivo experiments are required to determine whether the tumour-initiating cells are comparable to the in vitro phenotype, this study provides evidence of the stem cell origin of prostate cancer and can identify new therapeutic targets.

The Role of Androgen Signalling in the Regulation of Human Prostate Epithelial Stem Cell Differentiation

Dr A.T. Collins, Mr P. Berry and Prof N.J. Maitland

Collaborators:
Mr M. Stower (York Hospital NHS Trust)

Androgen-sensitive stromal cells have long been suspected of having an effector role in the development of prostate cancer, in addition to normal prostate development. We have set out to establish the phenotype of stromal cultures from normal and malignant tissues, using standard immunocytochemical and fluorescence cell sorting (FACS) with monoclonal antibodies to quantify expression of androgen receptor (AR), vimentin, fibronectin, smooth muscle actin, procollagen 1, and b1 integrin proteins. As a result, isolation of b1 integrin-expressing cells is a straightforward method to enrich for the AR-positive stromal cell population, since fewer than 10% of the b1 positive cells do not express AR.

The time course of stromal AR protein expression was next established following treatment with the androgen dihydrotestosterone (DHT). Using this time course, we have determined the expression of putative paracrine factors in stromal cell samples in response to androgens. These factors were selected with reference to published data, including FGF2, FGF7/KGF, TGFb, EGF, Wnt-4, Gli-1,and patched and smoothened receptors of the hedgehog signalling pathway, all of which have been shown to be important for defining the cancer phenotype in other stem cell systems.

To extend these studies to unknown paracrine factors, enriched androgenic stromal cultures from normal and malignant tissues were treated with DHT for various times, and the extracted RNA subjected to Affymetrix transcriptional profiling.

Lastly, the effects of these factors on stem cell colony forming ability, proliferation and differentiation markers will also be assessed. Therapeutically, such protein effectors may offer a new opportunity to impose differentiation control over tumour stem cells, and to eliminate androgen-resistant prostate cancers.

Microarray genotyping and phenotyping of Prostate Cancer Stem Cells

Dr S.D. Bryce, Dr A.T. Collins and Prof. N.J. Maitland

Collaborators:
Mr M. Stower (York Hospital NHS Trust)

Expression profiling with prostate samples has used heterogeneous tumour tissue samples, which include considerable contamination with stromal cells, infiltrating lymphocytes and macrophages. Even microdissected samples are heterogeneous in terms of malignant epithelial cell differentiation states. We estimate that the cancer stem cells comprise only 0.01% of the cells in the tumour; a unique expression signature for this population would be impossible to derive from a heterogeneous sample. To overcome this we have used cell selection based on the expression of cell surface markers to derive highly enriched populations of cells. This strategy allowed comparison between stem cell populations from malignant and benign samples leading to a malignant expression signature and comparison between stem and committed basal cell populations to determine a stem cell expression signature.

In order to investigate the underlying genetic alterations in the same cells from which we derived expression profiles, we employed Single Nucleotide Polymorphism (SNP) array analysis and microsatellite genotyping. The microsatellite analysis targeted selected loci for which allele loss was frequently seen in our previous work. SNP arrays have the advantage of simultaneously genotyping up to 10,240 different loci distributed over the whole genome. Again the tumour cells were fractionated to overcome inherent problems of tissue heterogeneity. For this first analysis we used the Affymetrix 10K 2.0 SNP arrays (Xba142 GeneChips). In both instances genotype data for genomic DNA from tumour derived cells were compared to genomic DNA from peripheral blood lymphocytes from the same patient. Loss of heterozygosity (LOH) was clearly detected at SNP loci, which are heterozygous in the matched normal sample, but was only been seen extensively with tissue from a lymph node metastasis. Local regions of amplification can also predicted from the SNP array data but this is more difficult to quantify.

Alterations in the activation pathway of Protein Kinase B in the progression to metastasis of human prostate cancer

Dr R.M. Sharrard, Ms J. Spalton and Prof. N.J. Maitland

Collaborators:
Dr V. Knauper, Dr H. Isaacs, Dr M. Rumsby and Dr B. Pownall (University of York)
Prof. D. Alessi and Prof. C.P. Downes (University of Dundee)
Prof. C. Cooper and Dr J. Clark (Institute of Cancer Research, Sutton, Surrey)
Dr V. Macaulay and Dr C. McCarthy (University of Oxford)

Development of invasiveness and metastasis in carcinomas is characterised by cell growth and survival independent of growth factors (GFs) and cell-cell and cell-matrix contact. These functions are modulated by Protein Kinase B (PKB), which is activated by phosphatidylinositol 3,4,5-trisphosphate generated by phosphatidylinositol 3-kinase (PI3K). PKB is critical in maintaining normal epithelial cell behaviour in a wide variety of tissues, including prostate. Alterations to PKB activation mechanisms are likely to be central to deregulated proliferation, inappropriate phenotypic development, and altered survival potential characteristics typical of tumour progression. PKB has been implicated in the maintenance of the stem cell phenotype; deregulation of its activity may support the emergence of stem cell-like properties within tumour cell populations.

Enhanced survival of metastatic prostate tumour cells has been linked to PKB deregulation through loss of the tumour suppressor protein PTEN. However, cell lines derived from prostate cancer metastases show additional changes to their mechanism of PKB activation in response to serum, GF stimulation or attachment to growth substrate compared to non-tumour or early tumour-derived cell lines.

We investigated the mechanisms of PKB activation in tumour and non-tumour cells using specific inhibitors of kinases of the EGFR/erbB, src, and PI3K families. SiRNA-directed knockdown of expression of IGF1 receptor, the transducers IRS1/2, and specific PI3K isoforms was used in combination with the inhibitors to elucidate the roles of these proteins in EGF and IGF1 signal transduction. Quantitation of the effects of these treatments allowed us to determine patterns of molecular changes responsible for altered response to GFs in prostate cancer progression. Using panels of kinase inhibitors to profile alterations in PKB signalling pathways in cells grown from tumour biopsies will have important clinical applications, allowing improved prognosis and the use of specific combinations of kinase inhibitor drugs for selective targeting of prostate tumour cells for destruction.


Yorkshire Cancer Research P53 Research Group
Director of Unit: Professor A.J. Milner

P53 post-translational modifications and their influence on histone H3 modifications
Dr. R.E. Adamson, Dr. S.J. Allison and Prof. J. Milner

We have transiently expressed human and murine wtp53 and p53 phosphorylation mutant proteins in both p53 null (-/-) human and murine cells. By Western blotting, we examined how alterations in p53 N-terminal phosphorylation status effect p53 C-terminal acetylation profiles and changes in p53 target gene expression. Point mutations at S15 and S37 result in enhanced phosphorylation at neighbouring S33 and S46 residues and at S392. Mutations at these residues increase C-terminal acetylation and enhance Ab421 binding. Furthermore, enhanced C-terminal acetylation results in increased p21 and MDM2 expression. Thus, N- terminal phosphorylation events play integral roles in p53 transcription factor function. We also conclude that p53 modifications are species specific, since cellular context did not influence the experimental outcomes.

Next, we examined the role of the p53 phosphorylation on histone modifications. P53 regulates chromatin-related processes. Under stress conditions, p53 activates the expression of genes necessary for DNA repair, cell growth arrest and apoptosis. It interacts with histone modifying complexes, influencing histone modifications. Mutating S15 to A or D and S37 to A, results in enhanced K4 trimethylation, K9 and K14 acetylation, and S10 phosphorylation on H3. Thus, we have uncovered a direct link between p53 function and H3 modifications.

Investigating novel genes within the SIRT1/FoxO4 apoptotic pathway
Dr S.U. Ahmed and Prof. J. Milner

SIRT1 is an NAD-dependent protein deacetylase which functions to enable human epithelial cancer cell growth and survival, (Ford et al. Cancer Research. 65(22): 2005). SIRT1 silencing induced growth arrest and/or apoptosis of the epithelial cancer cell lines where as normal human epithelial and diploid fibroblasts were refractory to SIRT1 silencing. Co-silencing experiments indicated that FoxO4, a member of the forkhead transcription factor, was an essential mediator of the SIRT1 apoptotic pathway.

We have investigated key differences in the function of these genes in tumour versus normal cell lines using RNAi. This powerful technique has allowed us to investigate apoptotic pathways with exquisite accuracy under conditions of minimal applied stress, and has been successfully applied in the identification of the Bcl-2/p53 pathway, (Jiang M and Milner J, Genes Dev. 17(7): 2003). This pathway is independent of the SIRT1/FoxO4 pathway, and this is particularly exciting as both apoptotic pathways can both operate in the same lineage.

Here, we present data identifying novel genes involved in the SIRT1/FoxO4 apoptotic pathway. The discovery of genes that operate specifically in tumour cell lines will aid the development of new anti-cancer therapies, which selectively kill cancer versus non-cancer cells.

Chromatin remodelling and cancer - regulation of histone H3 modification by the tumour suppressor p53 and perturbation of histone modification by the viral oncogene HPV E7
Dr. S.J. Allison and Prof. J. Milner

The packaging of DNA into chromatin is a major barrier to transcription, DNA repair, recombination and replication. Chromatin remodelling, induced by ATP-dependent remodelling complexes or post-translational modification of nucleosomal histones, has a key role in the control of such chromatin-related processes and is commonly subverted in cancers. Our research has focused on the contribution of chromatin remodelling to p53 function with particular emphasis on histone modification. Recent research has also been investigating the possible role of histone modification in subversion of host cell controls by the viral oncogene HPV E7.

Cancer-specific functions of SIRT1 enables human epithelial cancer cell growth and surbvival
Dr. J.R. Ford, Dr. M. Jiang and Prof. J. Milner

SIRT is a conserved NAD-dependent deacetylase, which regulates lifespan in accord with nutritional provision. In mammalian cells SIRT1 also down-regulates stress-induced p53- and FoxO pathways for apoptosis, thus favouring survival under stress. The functioning of SIRT1 under normal, non-stressed conditions of cell growth is unknown. Here we have asked if SIRT1 has the capacity to influence cell viability in the absence of applied stress. For this purpose we employed synthetic siRNA to silence SIRT1 gene expression by RNA interference (RNAi). We show that the process of RNA interference, by itself, does not affect growth and is not sufficient to activate a cellular stress response (indicated by lack of activation of endogenous p53). We also demonstrate that, in the absence of applied stress, SIRT1 silencing induces growth arrest and/or apoptosis in human epithelial cancer cells and normal human diploid fibroblasts appear to be refractory to SIRT1 silencing. Combined gene knock-out with RNAi co-silencing experiments indicate that SIRT1 and Bcl-2 may suppress separable apoptotic pathways in the same cell lineage and that the SIRT1-regulated pathway is independent of p53, of Bax and of caspase 2. Alternatively, SIRT1 may suppress apoptosis downstream from these apoptotic factors. In either case we show that FoxO4 (but not FoxO3) is required as pro-apoptotic mediator. We further identify caspases 3 and 7 as downstream executioners of SIRT1/FoxO4-regulated apoptosis. Our work identifies SIRT1 as a novel target for selective killing of cancer versus non-cancer epithelial cells.

A bi-functional siRNA construct induces RNA interference and also primes PCR amplification for its own quantification
Dr. M. Jiang and Prof. J. Milner

RNA interference (RNAi) is a process of post-transcriptional gene silencing initiated by double-stranded RNAs, including short interfering RNA (siRNA). Silencing is sequence-specific and RNAi has rapidly become central to the study of gene function. RNAi also carries promise for selective silencing of viral and endogenous genes causal for disease. To detect the very low levels of siRNA effective for RNAi we modified the 3'-end of the sense strand of siRNA with a nuclease-resistant DNA hairpin. We show that the modified siRNA-DNA construct (termed 'crook' siRNA) functions as a primer for the PCR and describe a novel, yet simple PCR protocol for its quantification (amolar levels/cell). When transfected into mammalian cells, crook siRNA induces selective mRNA knock-down equivalent to its unmodified siRNA counterpart. This new bi-functional siRNA construct will enable future in vivo studies on the uptake, distribution and pharmacokinetics of siRNA, and is particularly important for the development of siRNA-based therapeutics. More generally, PCR-based detection of siRNA carries wide-ranging applications for RNAi reverse genetics.

Investigation of apoptosis pathway regulated by HPV in cervical cancer
Dr. M. Jiang, Mr. D. Kidd and Prof. J. Milner

Cervical cancer is caused by HPV infection. It is the second most common cancer in women and causes over quarter of a million people to die all over the world each year, especially in developing countries. The newly developed vaccines have been proven to prevent initial viral infections. But it can only prevent the viral infection, not cure it. Under the current circumstances, we are still facing over 10 million deaths due to the existing HPV infections and its latency is up to 40 years. Thus there is an urgent need to obtain a thorough understanding of the molecular biology of HPV infections and to discover new targets and drugs for effective treatment of HPV infections.

Our previous study has shown that the synthetic siRNA can specifically knock down E7 expression and induce apoptosis in cervical cancer cell line, SiHa. Because E6 and E7 integrated into human genome as a bicistronic, we designed further four different siRNAs, which targets various regions of E6/7 bicistronic DNA. We found only the siRNAs targeting E7 encoding region can induce apoptosis. Co-silencing E7 siRNA with p53 siRNA or Rb siRNA shows that the apoptosis is p53 and Rb independent. In addition, a novel E6/E7 transcript from cervical cancer cells, named Sevix, has been identified. Sevix can partially impair the E7 regulated apoptosis and is also involved in the regulation of cellular DNA replication. This new transcript may also be a crucial player to control cell survival in HPV related cancers. In addition, using splicing junction specific siRNA to specific silencing individual transcripts, we have found that the individual transcript from the E6/E7 bicistronic DNA may have different functions towards the cell growth and apoptosis regulation. Function analysis of all four E6/E7 transcripts expressed in NHEK cells is undergoing.

Loss of one p53 allele results in four-fold reduction of p53 mRNA and protein: a basis for p53 haplo-insufficiency
Mr. C. Lynch and Prof. J. Milner

A haploid genotype may be insufficient to support normal wild-type function. Such haplo-insufficiency has recently been documented for numerous tumour suppressor genes. p53 is a crucial tumour suppressor governing DNA repair, cell cycle arrest and apoptosis via its role as a stress-responsive transcription factor. p53 haplo-insufficiency has been observed in vivo with human familial cancer in Li-Fraumeni Syndrome (LFS) and in mouse p53-knockout models of LFS. The increased tumorigenesis associated with loss of one p53 allele has been attributed to reduced p53-dependent stress responses. However the underlying biochemical basis for such attenuated responses in p53+/- cells remains unclear. Here we have determined basal p53 mRNA and protein levels, and compared the p53 stress response in p53+/+, p53+/- and p53-/- isogenic clones derived from HCT116 cells. Basal expression of p53 in p53+/- cells was 25% relative to p53+/+ cells, and this differential was maintained following oncogenic stress. This deficiency was manifest at both p53 mRNA and protein levels and resulted in attenuated p53 stress responses, in particular for p21waf1 trans-activation and Survivin trans-repression, with concurrent reduced G1 arrest and apoptosis.

Conclusions: These observations identify a molecular basis for wild-type p53 haplo-insufficiency, which may explain the attenuated tumour suppressive phenotype observed in cells with a single wild-type p53 allele and in humans with LFS.

Targeting the nucleolus for non-DNA damage-based anti-cancer therapy
Dr. C.P. Rubbi and Prof. J. Milner

Most anti-cancer therapies are based on DNA damage, which has the serious drawbacks of high mutagenicity and, due to the involvement of the p53 protein, secondary effects of the non-specific activation of p53 in normal cells. We recently discovered that disruption of the nucleolus causes p53 stabilisation by loss of the cellular capacity to degrade it, a finding welcomed by cancer researchers seeking ways to activate p53 for apoptosis without DNA damage and with high cancer specificity. This pathway of p53 activation is used by many cellular stresses, including those used for therapy. With 350 nucleolar proteins identified by proteomic analysis, there is now a chance to find cancer-specific targets for induction of apoptosis without DNA damage. This project will develop the nucleolar model of p53 activation by identifying the nucleolar functions that impinge on p53 homeostasis and nucleolar proteins that can be potential targets for anti-cancer therapy.

Role of phosphorylation in p53 acetylation and Pab421 epitope recognition in baculoviral and mammalian expressed proteins
Dr. L.J. Warnock, Miss S.A. Raines and Prof. J. Milner

It is current opinion that post-translational modifications such as phosphorylation and acetylation of the tumour suppressor protein p53, elicit important effects on the function and the stability of the resultant protein. However, as phosphorylation and acetylation are dynamic events subject to complex controls, elucidating the relationships between phosphorylation and acetylation is difficult. In the present study we have sought to address this problem by comparing full-length wild type p53 and full-length p53 proteins mutated at specific phosphorylation targets. Recombinant murine p53 proteins were expressed in insect cells using the baculoviral expression vector system (BVEs) and in a mammalian in vitro transcription/translation reticulocyte lysate system. In p53 proteins derived from BVEs, S37A (but not S37D) abrogates phosphorylation at S15. Lysine 382 (K382) is constitutively acetylated and shown to form part of the epitope recognised by PAb421. Lysine 373 (K373) is only acetylated following substitutions at S315 (S315A or S315D) or at S378 (S378A). Importantly, in baculoviral expressed proteins, PAb421 reactivity was independent of K373 acetylation status, indicating that acetylation at K382 specifically determines the PAb421 epitope.