SIX3 and SIX6 interact with GEMININ via C-terminal regions.

The histoarchitecture and function of eye and forebrain depend on a well-controlled balance between cell proliferation and differentiation. For example, the binding of the cell cycle regulator GEMININ to CDT1, which is a part of the pre-replication complex, promotes cell differentiation. Homeodomain transcription factors SIX3 and SIX6 also interact with GEMININ of which SIX3-GEMININ interaction promotes cell proliferation, whereas the nature of SIX6-GEMININ interaction has not been studied to date. We investigated SIX3/SIX6 and GEMININ interactions using bimolecular fluorescence complementation, surface plasmon resonance and isothermal titration calorimetry. Interactions between SIX3/SIX6 and GEMININ were detected in mammalian cells in culture. The presence of the C-terminal regions of SIX3 and SIX6 proteins, but not their SIX domains or homeodomains as previously thought, were required for interaction with GEMININ. Interestingly, the disordered C- and N- terminal regions of GEMININ were involved in binding to SIX3/SIX6. The coiled-coil region of GEMININ, which is the known protein-binding domain and also interacts with CDT1, was not involved in GEMININ-SIX3/SIX6 interaction. Using SPR and ITC, SIX3 bound GEMININ with a micromolar affinity and the binding stoichiometry was 1:2 (SIX3 - GEMININ). The present study gives new insights into the binding properties of SIX proteins, especially the role of their variable and disordered C-terminal regions.

Two tales of Annexin A2 knock-down: One of compensatory effects by antisense RNA and another of a highly active hairpin ribozyme.

Besides altering its own expression during cell transformation, Annexin A2 is upregulated during the progression of many cancer types and also plays key roles during viral infection and multiplication. Consequently, there has been great interest in Annexin A2 as a potential drug target. The successful design of efficient in vivo delivery systems constitutes an obstacle in full exploitation of antisense and RNA-cleaving technologies for the knock-down of specific targets. Efficiency is dependent on the method of delivery and accessibility of the target. Here, hairpin ribozymes and an antisense RNA against rat annexin A2 mRNA were tested for their efficiencies in a T7-driven coupled transcription/translation system. The most efficient ribozyme and antisense RNA were subsequently inserted into a retroviral vector under the control of a tRNA promoter, in a cassette inserted between retroviral Long Terminal Repeats for stable insertion into host DNA. The Phoenix package system based on defective retroviruses was used for virus-mediated gene transfer into PC12 cells. Cells infected with the ribozyme-containing particles died shortly after infection. However, the same ribozyme showed a very high catalytic effect in vitro in cell lysates, explained by its loose hinge helix 2 region. This principle can be transferred to other ribozymes, such as those designed to cleave the guide RNA in the CRISPR/Cas9 technology, as well as to target specific viral RNAs. Interestingly, efficient down-regulation of the expression of Annexin A2 by the antisense RNA resulted in up-regulation of Annexin A7 as a compensatory effect after several cell passages. Indeed, compensatory effects have previously been observed during gene knock-out, but not during knock-down of protein expression. This highlights the problems in interpreting the phenotypic effects of knocking down the expression of a protein. In addition, these data are highly relevant when considering the effects of the CRISPR/Cas9 approach.

CorA is a copper repressible surface-associated copper(I)-binding protein produced in Methylomicrobium album BG8.

CorA is a copper repressible protein previously identified in the methanotrophic bacterium Methylomicrobium album BG8. In this work, we demonstrate that CorA is located on the cell surface and binds one copper ion per protein molecule, which, based on X-ray Absorption Near Edge Structure analysis, is in the reduced state (Cu(I)). The structure of endogenously expressed CorA was solved using X-ray crystallography. The 1.6 Å three-dimensional structure confirmed the binding of copper and revealed that the copper atom was coordinated in a mononuclear binding site defined by two histidines, one water molecule, and the tryptophan metabolite, kynurenine. This arrangement of the copper-binding site is similar to that of its homologous protein MopE* from Metylococcus capsulatus Bath, confirming the importance of kynurenine for copper binding in these proteins. Our findings show that CorA has an overall fold similar to MopE, including the unique copper(I)-binding site and most of the secondary structure elements. We suggest that CorA plays a role in the M. album BG8 copper acquisition.

CXXC5 (retinoid-inducible nuclear factor, RINF) is a potential therapeutic target in high-risk human acute myeloid leukemia.

The retinoid-responsive gene CXXC5 localizes to the 5q31.2 chromosomal region and encodes a retinoid-inducible nuclear factor (RINF) that seems important during normal myelopoiesis. We investigated CXXC5/RINF expression in primary human acute myeloid leukemia (AML) cells derived from 594 patients, and a wide variation in CXXC5/RINF mRNA levels was observed both in the immature leukemic myeloblasts and in immature acute lymphoblastic leukemia cells. Furthermore, patients with low-risk cytogenetic abnormalities showed significantly lower levels compared to patients with high-risk abnormalities, and high RINF/CXXC5/ mRNA levels were associated with decreased overall survival for patients receiving intensive chemotherapy for newly diagnosed AML. This association with prognosis was seen both when investigating (i) an unselected patient population as well as for patients with (ii) normal cytogenetic and (iii) core-binding factor AML. CXXC5/RINF knockdown in AML cell lines caused increased susceptibility to chemotherapy-induced apoptosis, and regulation of apoptosis also seemed to differ between primary human AML cells with high and low RINF expression. The association with adverse prognosis together with the antiapoptotic effect of CXXC5/RINF suggests that targeting of CXXC5/RINF should be considered as a possible therapeutic strategy, especially in high-risk patients who show increased expression in AML cells compared with normal hematopoietic cells.

Low BRAF and NRAS expression levels are associated with clinical benefit from DTIC therapy and prognosis in metastatic melanoma.

Metastatic melanoma is characterized by a poor response to chemotherapy. Furthermore, there is a lack of established predictive and prognostic markers. In this single institution study, we correlated mutation status and expression levels of BRAF and NRAS to dacarbazine (DTIC) treatment response as well as progression-free and overall survival in a cohort of 85 patients diagnosed with advanced melanoma. Neither BRAF nor NRAS mutation status correlated to treatment response. However, patients with tumors harboring NRAS mutations had a shorter overall survival (p < 0.001) compared to patients with tumors wild-type for NRAS. Patients having a clinical benefit (objective response or stable disease at 3 months) on DTIC therapy had lower BRAF and NRAS expression levels compared to patients progressing on therapy (p = 0.037 and 0.003, respectively). For BRAF expression, this association was stronger among patients with tumors wild-type for BRAF (p = 0.005). Further, low BRAF as well as NRAS expression levels were associated with a longer progression-free survival in the total population (p = 0.004 and <0.001, respectively). Contrasting low NRAS expression levels, which were associated with improved overall survival in the total population (p = 0.01), low BRAF levels were associated with improved overall survival only among patients with tumors wild-type for BRAF (p = 0.013). These findings indicate that BRAF and NRAS expression levels may influence responses to DTIC as well as prognosis in patients with advanced melanoma.

Functional characterisation of p53 mutants identified in breast cancers with suboptimal responses to anthracyclines or mitomycin.

BACKGROUND: Approximately 4300 different TP53 mutations have been reported in human cancers. TP53 mutations, in particular those affecting the L2/L3 domains, are associated with resistance to anthracycline or mitomycin treatment in breast cancer patients. While many mutations have been characterised functionally, novel TP53 mutations are continuously reported. Here, we characterise 10 p53 protein variants encoded by mutated TP53 (5 within and 5 outside L2/L3) detected in locally advanced or metastatic breast cancers. Each tumour was previously characterised for response to therapy, allowing comparison between in vivo and in vitro findings. METHODS: Mutated p53 variants were analysed for their ability to oligomerise with the wild-type protein and their subcellular localisation by immunoprecipitation and immunofluorescence, respectively. Their ability to induce transcription of target genes was determined by qPCR. Cellular growth rate, apoptosis and senescence were monitored by WST-1, TUNEL and beta-galactosidase assays, respectively. RESULTS: Immunoprecipitation assays revealed each mutant protein to retain binding capacity for wild-type p53, thus potentially acting in a dominant negative manner. Even though each p53 variant located predominantly in the nucleus, the percentage of cells with only nuclear p53 localisation varied between 60% and 90%. None of the p53 variants were able to induce target genes to levels similar to wild-type p53, nor where they able to reduce cellular growth rate, induce apoptosis or senescence similar to wild-type p53 after anthracycline treatment in vitro. CONCLUSIONS: All the 10 variants studied displayed inferior p53 functionality compared to the wild-type protein. GENERAL SIGNIFICANCE: Our data add further information characterising the effects of somatic TP53 mutations on p53 protein function and anthracycline resistance in breast cancer.

The Methylococcus capsulatus (Bath) secreted protein, MopE*, binds both reduced and oxidized copper.

Under copper limiting growth conditions the methanotrophic bacterium Methylococcus capsulatus (Bath) secrets essentially only one protein, MopE*, to the medium. MopE* is a copper-binding protein whose structure has been determined by X-ray crystallography. The structure of MopE* revealed a unique high affinity copper binding site consisting of two histidine imidazoles and one kynurenine, the latter an oxidation product of Trp130. In this study, we demonstrate that the copper ion coordinated by this strong binding site is in the Cu(I) state when MopE* is isolated from the growth medium of M. capsulatus. The conclusion is based on X-ray Near Edge Absorption spectroscopy (XANES), and Electron Paramagnetic Resonance (EPR) studies. EPR analyses demonstrated that MopE*, in addition to the strong copper-binding site, also binds Cu(II) at two weaker binding sites. Both Cu(II) binding sites have properties typical of non-blue type II Cu (II) centres, and the strongest of the two Cu(II) sites is characterised by a relative high hyperfine coupling of copper (A(||) =20 mT). Immobilized metal affinity chromatography binding studies suggests that residues in the N-terminal part of MopE* are involved in forming binding site(s) for Cu(II) ions. Our results support the hypothesis that MopE plays an important role in copper uptake, possibly making use of both its high (Cu(I) and low Cu(II) affinity properties.

Human protein N-terminal acetyltransferase hNaa50p (hNAT5/hSAN) follows ordered sequential catalytic mechanism: combined kinetic and NMR study.

N(α)-acetylation is a common protein modification catalyzed by different N-terminal acetyltransferases (NATs). Their essential role in the biogenesis and degradation of proteins is becoming increasingly evident. The NAT hNaa50p preferentially modifies peptides starting with methionine followed by a hydrophobic amino acid. hNaa50p also possesses N(ε)-autoacetylation activity. So far, no eukaryotic NAT has been mechanistically investigated. In this study, we used NMR spectroscopy, bisubstrate kinetic assays, and product inhibition experiments to demonstrate that hNaa50p utilizes an ordered Bi Bi reaction of the Theorell-Chance type. The NMR results, both the substrate binding study and the dynamic data, further indicate that the binding of acetyl-CoA induces a conformational change that is required for the peptide to bind to the active site. In support of an ordered Bi Bi reaction mechanism, addition of peptide in the absence of acetyl-CoA did not alter the structure of the protein. This model is further strengthened by the NMR results using a catalytically inactive hNaa50p mutant.

Receptor for activated protein C kinase 1 (RACK1) is overexpressed in papillary thyroid carcinoma.

BACKGROUND: The receptor for activated C kinase 1 (RACK1) has been shown to be overexpressed in several types of cancers such as breast, colon, melanomas, and lung. RACK1 is linked to Ras-Raf-mediated signal transduction and transformed foci formation of 3T3 cells in vitro, and since this pathway is central in papillary thyroid carcinoma (PTC) oncogenesis, we hypothesized that RACK1 could play a role in the development or maintenance of PTC. No report on RACK1 expression in thyroid tissue is available; the present study was therefore aimed at identifying possible correlation of RACK1 expression at the mRNA or protein level in normal thyroid tissue compared to PTC. METHODS: We used TaqMan quantitative reverse transcriptase-polymerase chain reaction and immunohistochemistry to study the RACK1 gene and protein expression in matched tumor and nontumor samples from 59 PTC patients. The tumor samples were divided into two main categories, low-risk (group 1-3) and high-risk (group 4-6), in accordance with both histological classification and clinical appearance. RESULTS: RACK1 mRNA and protein levels were found highly overexpressed in tumor samples, whereas Ki-Ras mRNA was found to be relatively unchanged. B-Raf mRNA expression was low and detected only in tumor samples. Sequencing analysis detected no mutations in RACK1 or Ki-Ras, but 62.7% of the patients harbored the B-Raf single-nucleotide substitution T1799A (codon V600E). Phosphorylated extracellular signal-regulated kinase (pERK) immunohistochemistry analysis demonstrated activation of the mitogen-activated protein kinase (MAPK) pathway in tumor cells. Poorly differentiated and undifferentiated PTCs expressed significantly higher RACK1 mRNA levels than well-differentiated PTCs (p<0.017). CONCLUSIONS: Taken together, our findings point to an important role of RACK1 protein in PTC development and progression. Our data also emphasize the importance of assessing protein expression and not only mRNA levels.

The human N-alpha-acetyltransferase 40 (hNaa40p/hNatD) is conserved from yeast and N-terminally acetylates histones H2A and H4.

Protein N(α)-terminal acetylation (Nt-acetylation) is considered one of the most common protein modification in eukaryotes, and 80-90% of all soluble human proteins are modified in this way, with functional implications ranging from altered protein function and stability to translocation potency amongst others. Nt-acetylation is catalyzed by N-terminal acetyltransferases (NATs), and in yeast five NAT types are identified and denoted NatA-NatE. Higher eukaryotes additionally express NatF. Except for NatD, human orthologues for all yeast NATs are identified. yNatD is defined as the catalytic unit Naa40p (Nat4) which co-translationally Nt-acetylates histones H2A and H4. In this study we identified and characterized hNaa40p/hNatD, the human orthologue of the yeast Naa40p. An in vitro proteome-derived peptide library Nt-acetylation assay indicated that recombinant hNaa40p acetylates N-termini starting with the consensus sequence Ser-Gly-Gly-Gly-Lys-, strongly resembling the N-termini of the human histones H2A and H4. This was confirmed as recombinant hNaa40p Nt-acetylated the oligopeptides derived from the N-termini of both histones. In contrast, a synthetically Nt-acetylated H4 N-terminal peptide with all lysines being non-acetylated, was not significantly acetylated by hNaa40p, indicating that hNaa40p catalyzed H4 N(α)-acetylation and not H4 lysine N(ε)-acetylation. Also, immunoprecipitated hNaa40p specifically Nt-acetylated H4 in vitro. Heterologous expression of hNaa40p in a yeast naa40-Δ strain restored Nt-acetylation of yeast histone H4, but not H2A in vivo, probably reflecting the fact that the N-terminal sequences of human H2A and H4 are highly similar to each other and to yeast H4 while the N-terminal sequence of yeast H2A differs. Thus, Naa40p seems to have co-evolved with the human H2A sequence. Finally, a partial co-sedimentation with ribosomes indicates that hNaa40p co-translationally acetylates H2A and H4. Combined, our results strongly suggest that human Naa40p/NatD is conserved from yeast. Thus, the NATs of all classes of N-terminally acetylated proteins in humans now appear to be accounted for.

Using VAAST to identify an X-linked disorder resulting in lethality in male infants due to N-terminal acetyltransferase deficiency.

We have identified two families with a previously undescribed lethal X-linked disorder of infancy; the disorder comprises a distinct combination of an aged appearance, craniofacial anomalies, hypotonia, global developmental delays, cryptorchidism, and cardiac arrhythmias. Using X chromosome exon sequencing and a recently developed probabilistic algorithm aimed at discovering disease-causing variants, we identified in one family a c.109T>C (p.Ser37Pro) variant in NAA10, a gene encoding the catalytic subunit of the major human N-terminal acetyltransferase (NAT). A parallel effort on a second unrelated family converged on the same variant. The absence of this variant in controls, the amino acid conservation of this region of the protein, the predicted disruptive change, and the co-occurrence in two unrelated families with the same rare disorder suggest that this is the pathogenic mutation. We confirmed this by demonstrating a significantly impaired biochemical activity of the mutant hNaa10p, and from this we conclude that a reduction in acetylation by hNaa10p causes this disease. Here we provide evidence of a human genetic disorder resulting from direct impairment of N-terminal acetylation, one of the most common protein modifications in humans.

Predictive and prognostic impact of TP53 mutations and MDM2 promoter genotype in primary breast cancer patients treated with epirubicin or paclitaxel.

BACKGROUND: TP53 mutations have been associated with resistance to anthracyclines but not to taxanes in breast cancer patients. The MDM2 promoter single nucleotide polymorphism (SNP) T309G increases MDM2 activity and may reduce wild-type p53 protein activity. Here, we explored the predictive and prognostic value of TP53 and CHEK2 mutation status together with MDM2 SNP309 genotype in stage III breast cancer patients receiving paclitaxel or epirubicin monotherapy. EXPERIMENTAL DESIGN: Each patient was randomly assigned to treatment with epirubicin 90 mg/m(2) (n = 109) or paclitaxel 200 mg/m(2) (n = 114) every 3rd week as monotherapy for 4-6 cycles. Patients obtaining a suboptimal response on first-line treatment requiring further chemotherapy received the opposite regimen. Time from last patient inclusion to follow-up censoring was 69 months. Each patient had snap-frozen tumor tissue specimens collected prior to commencing chemotherapy. PRINCIPAL FINDINGS: While TP53 and CHEK2 mutations predicted resistance to epirubicin, MDM2 status did not. Neither TP53/CHEK2 mutations nor MDM2 status was associated with paclitaxel response. Remarkably, TP53 mutations (p = 0.007) but also MDM2 309TG/GG genotype status (p = 0.012) were associated with a poor disease-specific survival among patients having paclitaxel but not patients having epirubicin first-line. The effect of MDM2 status was observed among individuals harbouring wild-type TP53 (p = 0.039) but not among individuals with TP53 mutated tumors (p>0.5). CONCLUSION: TP53 and CHEK2 mutations were associated with lack of response to epirubicin monotherapy. In contrast, TP53 mutations and MDM2 309G allele status conferred poor disease-specific survival among patients treated with primary paclitaxel but not epirubicin monotherapy.

Proteome-derived peptide libraries allow detailed analysis of the substrate specificities of N(alpha)-acetyltransferases and point to hNaa10p as the post-translational actin N(alpha)-acetyltransferase.

The impact of N(α)-terminal acetylation on protein stability and protein function in general recently acquired renewed and increasing attention. Although the substrate specificity profile of the conserved enzymes responsible for N(α)-terminal acetylation in yeast has been well documented, the lack of higher eukaryotic models has hampered the specificity profile determination of N(α)-acetyltransferases (NATs) of higher eukaryotes. The fact that several types of protein N termini are acetylated by so far unknown NATs stresses the importance of developing tools for analyzing NAT specificities. Here, we report on a method that implies the use of natural, proteome-derived modified peptide libraries, which, when used in combination with two strong cation exchange separation steps, allows for the delineation of the in vitro specificity profiles of NATs. The human NatA complex, composed of the auxiliary hNaa15p (NATH/hNat1) subunit and the catalytic hNaa10p (hArd1) and hNaa50p (hNat5) subunits, cotranslationally acetylates protein N termini initiating with Ser, Ala, Thr, Val, and Gly following the removal of the initial Met. In our studies, purified hNaa50p preferred Met-Xaa starting N termini (Xaa mainly being a hydrophobic amino acid) in agreement with previous data. Surprisingly, purified hNaa10p preferred acidic N termini, representing a group of in vivo acetylated proteins for which there are currently no NAT(s) identified. The most prominent representatives of the group of acidic N termini are γ- and ß-actin. Indeed, by using an independent quantitative assay, hNaa10p strongly acetylated peptides representing the N termini of both γ- and ß-actin, and only to a lesser extent, its previously characterized substrate motifs. The immunoprecipitated NatA complex also acetylated the actin N termini efficiently, though displaying a strong shift in specificity toward its known Ser-starting type of substrates. Thus, complex formation of NatA might alter the substrate specificity profile as compared with its isolated catalytic subunits, and, furthermore, NatA or hNaa10p may function as a post-translational actin N(α)-acetyltransferase.

The MDM2 promoter SNP285C/309G haplotype diminishes Sp1 transcription factor binding and reduces risk for breast and ovarian cancer in Caucasians.

MDM2 plays a key role in modulating p53 function. The MDM2 SNP309T > G promoter polymorphism enhances Sp1 binding and has been linked to cancer risk and young age at diagnosis although with conflicting evidence. We report a second MDM2 promoter polymorphism, SNP285G > C, residing on the SNP309G allele. SNP285C occurs in Caucasians only, where 7.7% (95% CI 7.6%-7.8%) of healthy individuals carry the SNP285C/309G haplotype. In vitro analyses reveals that SNP309G enhances but SNP285C strongly reduces Sp1 promoter binding. Comparing MDM2 promoter status among different cohorts of ovarian (n = 1993) and breast (n = 1973) cancer patients versus healthy controls (n = 3646), SNP285C reduced the risk of both ovarian (OR 0.74; CI 0.58-0.94) and breast cancer (OR 0.79; CI 0.62-1.00) among SNP309G carriers.

Alterations of the retinoblastoma gene in metastatic breast cancer.

Germline mutations affecting the retinoblastoma gene (RB1) predispose to inherited retinoblastomas but also other malignancies, including breast cancer. While somatic RB1 mutations have been detected in different malignancies, information about the potential role of RB1 mutations in breast cancer is limited. Recently, we discovered RB1 mutations to be associated with resistance to anthracyclines/mitomycin in primary breast cancer. The present work is the first report evaluating RB1 mutation and epigenetic status in metastatic breast cancer. Among 148 breast cancer samples analyzed by MLPA, four samples harbored intragenic deletions/duplications: Thus, exons 1-2 were deleted in two tumors and exons 21-23 in one tumor, while one sample harbored duplication of exons 18-23. The entire RB1 gene was duplicated in two tumors and multiple amplifications were revealed in one sample. Reduced copy number was observed in 17 samples (11.5%). No point mutation or promoter hypermethylation was discovered (n = 38 and 114 tumors analyzed, respectively). Interestingly, among seven tumors expressing lack of response to epirubicin, two samples harbored alterations in RB1, contrasting none out of 16 tumors with stable disease or an objective response (P = 0.08). In summary, the frequency of RB1 alterations in metastatic lesions was not increased when compared to primary breast cancer, indicating that RB1 alterations do not play a major role in metastatic development. While a non-significant association suggesting RB1 alterations to be linked to therapy resistance was observed, our data do not suggest a major role for RB1 alterations explaining acquired drug resistance.

Identification and characterization of retinoblastoma gene mutations disturbing apoptosis in human breast cancers.

BACKGROUND: The tumor suppressor pRb plays a key role regulating cell cycle arrest, and disturbances in the RB1 gene have been reported in different cancer forms. However, the literature reports contradictory findings with respect to a pro--versus anti--apoptotic role of pRb, and the consequence of alterations in RB1 to chemotherapy sensitivity remains unclear. This study is part of a project investigating alterations in pivotal genes as predictive factors to chemotherapy sensitivity in breast cancer. RESULTS: Analyzing 73 locally advanced (stage III) breast cancers, we identified two somatic and one germline single nucleotide changes, each leading to amino acid substitution in the pRb protein (Leu607Ile, Arg698Trp, and Arg621Cys, respectively). This is the first study reporting point mutations affecting RB1 in breast cancer tissue. In addition, MLPA analysis revealed two large multiexon deletions (exons 13 to 27 and exons 21 to 23) with the exons 21-23 deletion occurring in the tumor also harboring the Leu607Ile mutation. Interestingly, Leu607Ile and Arg621Cys point mutations both localize to the spacer region of the pRb protein, a region previously shown to harbor somatic and germline mutations. Multiple sequence alignment across species indicates the spacer to be evolutionary conserved. All three RB1 point mutations encoded nuclear proteins with impaired ability to induce apoptosis compared to wild-type pRb in vitro. Notably, three out of four tumors harboring RB1 mutations displayed primary resistance to treatment with either 5-FU/mitomycin or doxorubicin while only 14 out of 64 tumors without mutations were resistant (p = 0.046). CONCLUSIONS: Although rare, our findings suggest RB1 mutations to be of pathological importance potentially affecting sensitivity to mitomycin/anthracycline treatment in breast cancer.

MGMT expression levels predict disease stabilisation, progression-free and overall survival in patients with advanced melanomas treated with DTIC.

Metastatic melanoma responds poorly to systemic treatment. We report the results of a prospective single institution study evaluating O(6)-methylguanine-DNA methyltransferase (MGMT) status as a potential predictive and/or prognostic marker among patients treated with dacarbazine (DTIC) 800-1000 mg/m(2) monotherapy administered as a 3-weekly schedule for advanced malignant melanomas. The study was approved by the Regional Ethical Committee. Surgical biopsies from metastatic or loco-regional deposits obtained prior to DTIC treatment were snap-frozen immediately upon removal and stored in liquid nitrogen up to processing. Median time from enrolment to end of follow-up was 67 months. MGMT expression levels evaluated by qRT-PCR correlated significantly to DTIC benefit (CR/PR/SD; p=0.005), time to progression (TTP) (p=0.005) and overall survival (OS) (p=0.003). MGMT expression also correlated to Breslow thickness in the primary tumour (p=0.014). While MGMT promoter hypermethylation correlated to MGMT expression, MGMT promoter hypermethylation did not correlate to treatment benefit, TTP or OS, suggesting that other factors may be critical in determining MGMT expression levels in melanomas. In a Cox proportional regression analysis, serum lactate dehydrogenase (LDH, p<0.001), MGMT expression (p=0.022) and p16(INK4a) expression (p=0.037) independently predicted OS, while TTP correlated to DTIC benefit after 6 weeks only (p=0.001). Our data reveal MGMT expression levels to be associated with disease stabilisation and prognosis in patients receiving DTIC monotherapy for advanced melanoma. The role of MGMT expression as a predictor to DTIC sensitivity versus a general prognostic factor in advanced melanomas warrants further evaluation.

Gene expression profiling-based identification of molecular subtypes in stage IV melanomas with different clinical outcome.

PURPOSE: The incidence of malignant melanoma is increasing worldwide in fair-skinned populations. Melanomas respond poorly to systemic therapy, and metastatic melanomas inevitably become fatal. Although spontaneous regression, likely due to immune defense activation, rarely occurs, we lack a biological rationale and predictive markers in selecting patients for immune therapy. EXPERIMENTAL DESIGN: We performed unsupervised hierarchical clustering of global gene expression data from stage IV melanomas in 57 patients. For further characterization, we used immunohistochemistry of selected markers, genome-wide DNA copy number analysis, genetic and epigenetic analysis of the CDKN2A locus, and NRAS/BRAF mutation screening. RESULTS: The analysis revealed four distinct subtypes with gene signatures characterized by expression of immune response, pigmentation differentiation, proliferation, or stromal composition genes. Although all subtypes harbored NRAS and BRAF mutations, there was a significant difference between subtypes (P < 0.01), with no BRAF/NRAS wild-type samples in the proliferative subtype. Additionally, the proliferative subtype was characterized by a high frequency of CDKN2A homozygous deletions (P < 0.01). We observed a different prognosis between the subtypes (P = 0.01), with a particularly poor survival for patients harboring tumors of the proliferative subtype compared with the others (P = 0.003). Importantly, the clinical relevance of the subtypes was validated in an independent cohort of 44 stage III and IV melanomas. Moreover, low expression of an a priori defined gene set associated with immune response signaling was significantly associated with poor outcome (P = 0.001). CONCLUSIONS: Our data reveal a biologically based taxonomy of malignant melanomas with prognostic effect and support an influence of the antitumoral immune response on outcome.

Chk2 splice variants express a dominant-negative effect on the wild-type Chk2 kinase activity.

While the majority of RNA transcripts from protein-encoding genes in the human genome are subject to physiological splicing, pathological splicing is increasingly reported in cancer tissue. Previously, we identified >90 different splice variants of Chk2, a gene encoding a serine/threonine kinase propagating the DNA damage signal by phosphorylating and activating several downstream substrates like p53, Cdc25A, and Cdc25C involved in cell cycle arrest and apoptosis. While alternative splice forms of other genes have been reported to exert a dominant-negative effect on the wild-type molecules, the function of Chk2 splice protein variants is still unclear. Here we evaluated the function of four Chk2 splice proteins for which mRNA splice variants were identified in human breast carcinomas. These splice variants were stably expressed as nuclear proteins. Two splice forms (Chk2Delta4 and Chk2del(2-3)) expressed kinase activity while variants Chk2Delta11 and Chk2isoI were essentially kinase inactive. Independent of intrinsic kinase activity, each splice variant impaired wild-type Chk2 activity through heterodimerization. Based on our findings, we suggest alternative splicing as a possible novel mechanism for repression of the Chk2 wild-type function.