Retinoblastoma: From genes to patient care.

Retinoblastoma is the most common paediatric neoplasm of the retina, and one of the earliest model of cancer genetics since the identification of the master tumour suppressor gene RB1. Tumorigenesis has been shown to be driven by pathogenic variants of the RB1 locus, but also genomic and epigenomic alterations outside the locus. The increasing knowledge on this "mutational landscape" is used in current practice for precise genetic testing and counselling. Novel methods provide access to pre-therapeutic tumour DNA, by isolating cell-free DNA from aqueous humour or plasma. This is expected to facilitate assessment of the constitutional status of RB1, to provide an early risk stratification using molecular prognostic markers, to follow the response to the treatment in longitudinal studies, and to predict the response to targeted therapies. The aim of this review is to show how molecular genetics of retinoblastoma drives diagnosis, treatment, monitoring of the disease and surveillance of the patients and relatives. We first recap the current knowledge on retinoblastoma genetics and its use in every-day practice. We then focus on retinoblastoma subgrouping at the era of molecular biology, and the expected input of cell-free DNA in the field.

Huntingtin with an expanded polyglutamine repeat affects the Jab1-p27(Kip1) pathway.

Expansion of polyglutamine repeats is the cause of at least nine inherited human neurodegenerative disorders, including Huntington's disease (HD). It is widely accepted that deregulation of the transcriptional coactivator CBP by expanded huntingtin (htt) plays an important role in HD molecular pathogenesis. In this study, we report on a novel target of expanded polyglutamine stretches, the transcriptional coactivator Jun activation domain-binding protein 1 (Jab1), which shares DNA-sequence-specific transcription factor targets with CBP. Jab1 also plays a major role in the degradation of the cyclin-dependent-kinase inhibitor and putative transcription cofactor p27(Kip1). We found that Jab1 accumulates in aggregates when co-expressed with either expanded polyglutamine stretches or N-terminal fragments of mutant htt. In addition, the coactivator function of Jab1 was suppressed both by aggregated expanded polyglutamine solely and by mutant htt. Inhibition by mutant htt even preceded the appearance of microscopic aggregation. In an exon 1 HD cell model, we found that endogenous Jab1 could be recruited into aggregates and that this was accompanied by the accumulation of p27(Kip1). Accumulation of p27(Kip1) was also found in brains derived from HD patients. The repression of Jab1 by various mechanisms coupled with an increase of p27(Kip1) at late stages may have important transcriptional effects. In addition, the interference with the Jab1-p27(Kip1) pathway may contribute to the observed lower incidence of cancer in HD patients and may also be relevant for the understanding of the molecular pathogenesis of polyglutamine disorders in general.

Distribution of inclusions in neuronal nuclei and dystrophic neurites in Huntington disease brain.

Recently, an N-terminal fragment of huntingtin was localized to neuronal intranuclear inclusions (NII), presumed to cause cellular dysfunction, and to inclusions in dystrophic neurites (IDN) in the neostriatum and neocortex of Huntington disease (HD) patients. In the present immunohistochemical study of autopsy brain of 2 juvenile-onset HD patients, 5 HD patients with adult-onset, and 5 controls, NII and IDN as stained with both N-terminal antiserum to huntingtin and ubiquitin antiserum were detected in the HD neostriatum, neocortex, and allocortex, but not in the HD pallidum, cerebellum, and substantia nigra nor in control brain. The frequency of NII in the HD neocortex was highest in the juvenile patients. Within the allocortex, NII and IDN were found in the entorhinal region, subiculum, and pyramidal cell layer of Ammon's horn. N-terminal huntingtin antiserum also labeled intranuclear granular structures adjacent to the neuronal nuclear membrane in 5 HD patients, one control with idiopathic epilepsy, and one with Alzheimer disease. Our results show that NII formation in HD involves the allocortex in addition to the neostriatum and neocortex. The development of NII in the neocortex and allocortex in HD brain might contribute to the emergence of the cognitive and behavioral symptoms of the disease.

Rubinstein-Taybi syndrome caused by a De Novo reciprocal translocation t(2;16)(q36.3;p13.3).

Rubinstein-Taybi syndrome (RTS) is a multiple congenital anomalies and mental retardation syndrome characterized by facial abnormalities, broad thumbs, and broad big toes. We have shown previously that disruption of the human CREB-binding protein (CBP) gene, either by gross chromosomal rearrangements or by point mutations, leads to RTS. Translocations and inversions involving chromosome band 16p13.3 form the minority of CBP mutations, whereas microdeletions occur more frequently (approximately 10%). Breakpoints of six translocations and inversions in RTS patients described thus far were found clustered in a 13-kb intronic region at the 5' end of the CBP gene and could theoretically only result in proteins containing the extreme N-terminal region of CBP. In contrast, in one patient with a translocation t(2;16)(q36.3;p13.3) we show by using fiber FISH and Southern blot analysis that the chromosome 16 breakpoint lies about 100 kb downstream of this breakpoint cluster. In this patient, Western blot analysis of extracts prepared from lymphoblasts showed both a normal and an abnormal shorter protein lacking the C-terminal domain, indicating expression of both the normal and the mutant allele. The results suggest that the loss of C-terminal domains of CBP is sufficient to cause RTS. Furthermore, these data indicate the potential utility of Western blot analysis as an inexpensive and fast approach for screening RTS mutations.

E1A + cHa-ras transformed rat embryo fibroblast cells are characterized by high and constitutive DNA binding activities of AP-1 dimers with significantly altered composition.

Transcription factors of the AP-1/ATF family, including c-Fos, c-Jun, and ATF-2, play an important role in the regulation of cell proliferation and differentiation, and changes in their levels and/or activities may contribute to oncogenesis. We analyzed the alterations of AP-1/ATF transcription factors upon immortalization and transformation in a panel of cell lines derived from rat embryo fibroblast (REF) cells. The tumorigenic E1A + cHa-ras cells are characterized by high and constitutive DNA binding activities of AP-1, in contrast to nontransformed cells and the E1A cells. The expression of c-fos and c-jun genes was affected differently by the oncogenic transformation. By using antibodies to c-Jun and c-Fos proteins in electrophoretic mobility shift assays (EMSA), we showed that E1A + cHa-ras transformants did not contain c-Fos under any condition of cell cultivation and growth factor stimulation, whereas c-Jun was constitutively upregulated. In the absence of c-fos gene expression, c-Fos protein appears to be replaced by proteins of Fos family (Fra-1) and ATF family (ATF-2 and ATFa). To determine the possible mechanisms of c-fos downregulation in E1A + cHa-ras transformants we have obtained populations of geneticin-resistant clones containing integrated reporter construct -711fos-CAT and its mutants in serum-responsive element (SRE) and cAMP-responsive element (CRE). Data obtained show that the mutations within the SRE lead to a manifold activation of fos-CAT expression. This allows to suggest that c-fos downregulation in E1A + cHa-ras transformants is provided by a negative control mediated through the SRE regulatory region. The profound differences in regulation and composition of transcription factors of the AP-1 family probably play a pivotal role in the transformation of REF cells by E1A and cHa-ras oncogenes.

HER-2/neu and p27Kip1 in progression of Fallopian tube carcinoma: an immunohistochemical and array comparative genomic hybridization study.

AIMS: To determine expression of p53, HER-2/neu and p27(Kip1) in serous Fallopian tube carcinoma (FTC) in relation to stage and grade, and to investigate DNA copy number changes of HER-2 and P27KIP1 as a potential mechanism of altered expression status. METHODS AND RESULTS: Immunohistochemistry was performed on 28 serous FTCs and 10 normal Fallopian tubes. p53 protein accumulated and p27(Kip1) was down-regulated significantly in early-stage FTCs compared with normal Fallopian tubes. HER-2/neu overexpression was absent in normal Fallopian tubes and in all stage I FTCs (n = 6) but present in 57% (12/21) of advanced-stage FTCs. No differences in expression between grade 2 and 3 tumours were detected. HER-2 gain/amplification was found by array comparative genomic hybridization in 23% (3/13) of analysed FTCs and all showed overexpression. HER-2/neu overexpression also occurred without DNA copy number changes in three other cases. For p27(Kip1), expression and DNA copy number were unrelated. CONCLUSIONS: p53 accumulation and p27(Kip1) down-regulation seem to be early events in Fallopian tube carcinogenesis. HER-2/neu showed overexpression, caused by gain/amplification in 50%, and may be involved in progression of FTC. These data contribute to a better understanding of the molecular carcinogenesis of FTC and to possible new therapeutic approaches.

DNA profiling of primary serous ovarian and fallopian tube carcinomas with array comparative genomic hybridization and multiplex ligation-dependent probe amplification.

Primary serous ovarian carcinoma (OVCA) and serous Fallopian tube carcinoma (FTC), both belonging to the BRCA-linked tumour spectrum, share many properties and are treated similarly. However, a detailed molecular comparison has been lacking. We hypothesized that comparative genomic studies of serous OVCAs and FTCs should point to gene regions critically involved in their tumorigenesis. Array comparative genomic hybridization (array CGH) analysis indicated that serous OVCAs and serous FTCs displayed common but also more distinctive patterns of recurrent changes. Targeted gene identification using a dedicated multiplex ligation-dependent probe amplification (MLPA) probe set directly identified EIF2C2 on 8q as a potentially important driver gene. Other previously unappreciated gained/amplified genes included PSMB4 on 1q, MTSS1 on 8q, TEAD4 and TSPAN9 on 12p, and BCAS4 on 20q. SPINT2 and ACTN4 on 19q were predominantly found in FTCs. Gains/amplifications of CCNE1 and MYC, often in conjunction with changes in genes of the AKT pathway, EVI1 and PTK2, seemed to be involved at earlier stages, whereas changes of ERBB2 were associated with advanced stages. The only BRCA1-mutated FTC shared common denominators with the sporadic tumours. In conclusion, the data suggest that serous OVCAs and FTCs, although related, exhibit differences in genomic profiles. In addition to known pathways, new genes/pathways are likely to be involved, with changes in an miRNA-associated gene, EIF2C2, as one important new feature. Dedicated MLPA sets constitute potentially important tools for differential diagnosis and may provide footholds for tailored therapy.

A case of loss of heterozygosity in the BRCA2 gene of a borderline ovarian tumor: case report and review of literature.

Germline BRCA1 and BRCA2 mutations highly increase the risk of breast and female adnexal cancer. The role of these genes in the tumorigenesis of other malignancies is still under debate. Borderline ovarian tumors (BOT) are occasionally found in families with a strong history of breast and/or female adnexal cancer with or without proven germline mutations. We investigated whether a BOT arising in a germline BRCA2 mutation carrier could be attributed to this mutation, in which case BOT should be added to the BRCA2 related tumor spectrum. Tumor DNA of a serous borderline ovarian tumor (sBOT) of a 55-year-old female carrier of a pathogenic BRCA2 mutation (6085G>T) was analyzed for loss of heterozygosity (LOH) of BRCA2. The sBOT cells, unexpectedly, revealed loss of the mutant allele of BRCA2, while ovarian stroma cells and peripheral blood lymphocytes contained both wild-type and mutant allele of BRCA2. The finding that no loss of the wild-type BRCA2 allele was found in the tumor tissue but loss of the mutant allele was seen suggests that sBOT are not part of the BRCA2 related tumor spectrum. In the literature BOT's in germline BRCA1 and BRCA2 mutation carriers are described incidentally, while in patients with a BOT a germline BRCA1 or BRCA2 mutation is rarely found. Therefore, we conclude that borderline ovarian tumors are neither part of the BRCA1- nor the BRCA2- related tumor spectrum.

Expression of the gene encoding subunit II of yeast QH2: cytochrome c oxidoreductase is regulated by multiple factors.

In Saccharomyces cerevisiae, the COR2 gene codes for the 40 kDa subunit II of the QH2: cytochrome c oxidoreductase, an enzyme of the mitochondrial respiratory chain. Regions in the 5' flank of this gene important for regulated expression were identified by assaying beta-galactosidase activities in cells carrying different COR2-lacZ fusion genes. Sequences downstream of position -201 relative to the translational initiation codon are sufficient to confer regulation by carbon source, whereas sequences downstream of position -153 do not give rise to significant expression. A binding site for the abundant general transcription factor GFI is present in the region between -201 and -153 just upstream from sequences which resemble the consensus DNA recognition sequence of the regulatory protein complex HAP2/HAP3: 5'-TNATTGGT-3'. By quantitating RNA levels and assaying beta-galactosidase activities we show that synthesis of COR2, which is not a hemoprotein, is regulated by HAP1, HAP2/HAP3 and heme.

Multi-functional DNA proteins in yeast: the factors GFI and GFII are identical to the ARS-binding factor ABFI and the centromere-binding factor CPF1 respectively.

GFI and GFII are abundant DNA-binding proteins in the yeast Saccharomyces cerevisiae. Binding sites for GFI conform to the sequence 5'-RTCRYNNNNNACG-3'. This consensus can also accommodate the recognition sequence for the ARS1-binding factor ABFI. Results of retardation-competition assays, limited proteolysis experiments, molecular weight determinations based on denaturation-renaturation procedures and mobility shift assays of protein-DNA complexes formed in the presence of a monoclonal antibody raised against ABFI suggest strongly that GFI and ABFI are the same protein. Similarly, GFII appears to be identical to the centromere-binding protein CPF1 (alias CP1), since both proteins bind to the CDEI motif of yeast centromeres (5'-RTCACRTG-3') and cannot be detected in a cpf1 disruption mutant yeast strain. In addition, based on denaturation-renaturation studies, both factors appear to have molecular weights in the same range of 53-62 kDa.

Identification of two factors which bind to the upstream sequences of a number of nuclear genes coding for mitochondrial proteins and to genetic elements important for cell division in yeast.

Two abundant factors, GFI and GFII which interact with the 5' flanking regions of nuclear genes coding for proteins of the mitochondrial respiratory chain have been identified. In one case (subunit VIII of QH2: cytochrome c oxidoreductase) the binding sites for both factors overlap completely and their binding is mutually exclusive. For the other 5' regions tested the GFI and GFII binding sites do not coincide. Interestingly, binding sites for GFI and GFII are also present in or at the 3' ends of the coding regions of two genes of the PHO gene family and in DNA elements important for optimal ARS and CEN function respectively. The sites recognized by GFI conform to the consensus RTCRNNNNNNACGNR, while those recognized by GFII contain the element RTCACGTG. We speculate that GFI and GFII may play a role in different cellular processes, dependent on the context of their binding sites and that one of these processes may be the coordination of the expression of genes involved in mitochondrial biogenesis with the progress of the cell cycle.

The adenovirus 12 E1A proteins can bind directly to proteins of the p300 transcription co-activator family, including the CREB-binding protein CBP and p300.

The cellular transcription co-activators p300 and the CREB-binding protein CBP are cellular targets for transformation by the E1A proteins of non-oncogenic adenovirus 5 (Ad5). In this study, we show that the E1A proteins of oncogenic Ad12, like those of Ad5, can also bind to CBP and that this interaction is direct. In addition, we show that the Ad12 E1A proteins can also bind directly to p300. These results suggest that E1A can modulate the function of proteins of the p300 family via direct protein-protein interactions.

Yeast general transcription factor GFI: sequence requirements for binding to DNA and evolutionary conservation.

GFI is an abundant DNA binding protein in the yeast S. cerevisiae. The protein binds to specific sequences in both ARS elements and the upstream regions of a large number of genes and is likely to play an important role in yeast cell growth. To get insight into the relative strength of the various GFI-DNA binding sites within the yeast genome, we have determined dissociation rates for several GFI-DNA complexes and found them to vary over a 70-fold range. Strong binding sites for GFI are present in the upstream activating sequences of the gene encoding the 40 kDa subunit II of the QH2:cytochrome c reductase, the gene encoding ribosomal protein S33 and in the intron of the actin gene. The binding site in the ARS1-TRP1 region is of intermediate strength. All strong binding sites conform to the sequence 5' RTCRYYYNNNACG-3'. Modification interference experiments and studies with mutant binding sites indicate that critical bases for GFI recognition are within the two elements of the consensus DNA recognition sequence. Proteins with the DNA binding specificities of GFI and GFII can also be detected in the yeast K. lactis, suggesting evolutionary conservation of at least the respective DNA-binding domains in both yeasts.

Subcellular localization of the Huntington's disease gene product in cell lines by immunofluorescence and biochemical subcellular fractionation.

Huntington's disease is a progressive neurodegenerative disorder, which is caused by expansion of a polymorphic (CAG)n repeat in the coding region of the Huntington's disease gene. The function of huntingtin has not been elucidated so far. Accordingly, detailed subcellular localization studies remain useful. In an immunohistochemical study, we have reported huntingtin to be present in the cytoplasm of cells in the majority of the tissues studied. In addition, we detected a signal in the nucleus of cells in some tissues, including neuronal cells. We have further extended these studies in various mammalian cell lines, using a panel of (affinity-purified) polyclonal huntingtin antibodies in immunofluorescence, confocal laser scanning microscopy and biochemical subcellular fractionation studies. In mouse embryonic fibroblasts, human skin fibroblasts and in mouse neuroblastoma cells huntingtin was present in the cytoplasm. All five antibodies, directed against different parts of huntingtin, also showed a signal in the nucleus. This signal could be competed by the original antigen. The localization of huntingtin in both cytoplasm and nucleus, was confirmed by biochemical subcellular fractionation studies. However, in most other studies, a nuclear location for huntingtin has not been found. Our results suggest, however, that besides its function(s) in the cytoplasm, a nuclear function of huntingtin at some stages of differentiation or in some phases of the cell cycle may not be excluded.

The adenoviral E1A oncoproteins interfere with the growth-inhibiting effect of the cdk-inhibitor p21(CIP1/WAF1).

The cdk-inhibitor p21(CIP1/WAF1) inhibits the activities of cyclin-dependent kinases and proliferating cell nuclear antigen, thereby repressing cell-cycle progression and DNA replication. Transforming oncogenes, such as E1A of human adenovirus 5 (Ad5), may interfere with such growth-inhibitory proteins. In this study, we show that in various Ad5E1-transformed cells, p21(CIP1/WAF1) is expressed and that, in general, expression is not downregulated. In addition, colony-formation assays show that in Ad5E1-transformed cells highly overexpressed p21(CIP1/WAF1) can still cause growth inhibition. FACS experiments indicate, however, that a G1 arrest induced by moderate overexpression of p21(CIP1/WAF1) can be overcome by E1A. The E1A proteins may interfere with the function of p21(CIP1/WAF1) by binding. Indeed, p21(CIP1/WAF1) binds with its cyclin/cdk-binding N terminus to the transforming N-terminal and CR1 region of the E1A proteins. Together, these results lend support to the model that E1A can interfere directly with p21(CIP1/WAF1) function and thereby stimulates cell growth.

The N-terminal region of the adenovirus type 5 E1A proteins can repress expression of cellular genes via two distinct but overlapping domains.

The transforming E1A 12S and E1A 13S proteins of human adenovirus type 5 (Ad5) contain two and three conserved regions, respectively. In the present study, the contribution of sequences in the nonconserved N-terminal region of the E1A proteins to morphological transformation and to down-regulation of a number of mitogen-inducible genes was investigated. As described previously, transformation of NRK cells (an established normal rat kidney cell line) results in denser cell growth and a cuboidal cellular morphology. None of the cells expressing N-terminally mutated E1A proteins, however, show these transformed properties, which suggests an important role for sequences in that domain. The decrease in cyclin D1 levels requires exactly the same sequences. The ability to transform NRK cells and to reduce cyclin D1 levels does not correlate with the presence in the E1A proteins of binding domains for p300, CBP, p107, pRb, cyclin A, or cdk2. In contrast, down-regulation of expression of the JE gene in NRK cells and repression of transcription of the collagenase gene in human HeLa cells does correlate with the presence in the E1A proteins of an intact binding domain for p300 and CBP. The results suggest that the N-terminal domain of the E1A proteins can repress expression of cellular genes by at least two different mechanisms.

An ARS/silencer binding factor also activates two ribosomal protein genes in yeast.

GFI is an abundant yeast DNA-binding protein, capable of binding to both ARS sequences and to the upstream regions of a number of nuclear genes coding for mitochondrial proteins (Dorsman et al., Nucl. Acids Res., 16 [1988] 7287-7301). GFI binding sites conform to the consensus RTCRYN5ACG, an element also present in the binding sites of factors designated SUF and TAF. These factors act as trans-activators of the constitutive transcription of the genes for ribosomal proteins S33 and L3 respectively. We now present evidence that GFI, TAF and SUF are probably the same protein. We speculate that one of the functions of GFI is the adjustment of the expression of a number of gene families to cell growth rate.

IkappaB alpha is a target for the mitogen-activated 90 kDa ribosomal S6 kinase.

The activity of transcription factor NFkappaB is regulated by its subcellular localization. In most cell types, NFkappaB is sequestered in the cytoplasm due to binding of the inhibitory protein IkappaB alpha. Stimulation of cells with a wide variety of agents results in degradation of IkappaB alpha which allows translocation of NFkappaB to the nucleus. Degradation of IkappaB alpha is triggered by phosphorylation of two serine residues, i.e. Ser32 and Ser36, by as yet unknown kinases. Here we report that the mitogen-activated 90 kDa ribosomal S6 kinase (p90rsk1) is an IkappaB alpha kinase. p90rsk1 phosphorylates IkappaB alpha at Ser32 and it physically associates with IkappaB alpha in vivo. Moreover, when the function of p90rsk1 is impaired by expression of a dominant-negative mutant, IkappaB alpha degradation in response to mitogenic stimuli, e.g. 12-O-tetradecanoylphorbol 13-acetate (TPA), is inhibited. Finally, NFkappaB cannot be activated by TPA in cell lines that have low levels of p90rsk1. We conclude that p90rsk1 is an essential kinase required for phosphorylation and subsequent degradation of IkappaB alpha in response to mitogens.

Analysis of the subcellular localization of huntingtin with a set of rabbit polyclonal antibodies in cultured mammalian cells of neuronal origin: comparison with the distribution of huntingtin in Huntington's disease autopsy brain.

Huntington's disease (HD) is a neurodegenerative disorder with a midlife onset. The disease is caused by expansion of a CAG (glutamine) repeat within the coding region of the HD gene. The molecular mechanism by which the mutated protein causes this disease is still unclear. To study the protein we have generated a set of rabbit polyclonal antibodies raised against different segments of the N-terminal, central and C-terminal parts of the protein. The polyclonal antibodies were affinity purified and characterized in ELISA and Western blotting experiments. All antibodies can react with mouse and human proteins. The specificity of these antibodies is underscored by their recognition of huntingtin with different repeat sizes in extracts prepared from patient-derived lymphoblasts. The antibodies were used in immunofluorescence experiments to study the subcellular localization of huntingtin in mouse neuroblastoma NIE-115 cells. The results indicate that most huntingtin is present in the cytoplasm, whereas a minor fraction is present in the nucleus. On differentiation of the NIE-115 cells in vitro, the subcellular distribution of huntingtin does not change significantly. These results suggest that full-length huntingtin with a normal repeat length can be detected in the nucleus of cycling and non-cycling cultured mammalian cells of neuronal origin. However, in HD autopsy brain the huntingtin-containing neuronal intranuclear inclusions can be detected only with antibodies raised against the N-terminus of huntingtin. Thus several forms of huntingtin display the propensity for nuclear localization, possibly with different functional consequences.