Propensities of Fatty Acid-Modified ASOs: Self-Assembly vs Albumin Binding.

We conducted a biophysical study to investigate the self-assembling and albumin-binding propensities of a series of fatty acid-modified locked nucleic acid (LNA) antisense oligonucleotide (ASO) gapmers specific to the MALAT1 gene. To this end, a series of biophysical techniques were applied using label-free ASOs that were covalently modified with saturated fatty acids (FAs) of varying length, branching, and 5'/3' attachment. Using analytical ultracentrifugation (AUC), we demonstrate that ASOs conjugated with fatty acids longer than C16 exhibit an increasing tendency to form self-assembled vesicular structures. The C16 to C24 conjugates interacted via the fatty acid chains with mouse and human serum albumin (MSA/HSA) to form stable adducts with near-linear correlation between FA-ASO hydrophobicity and binding strength to mouse albumin. This was not observed for the longer fatty acid chain ASO conjugates (>C24) under the experimental conditions applied. The longer FA-ASO however adopted self-assembled structures with increasing intrinsic stabilities proportional to the fatty acid chain length. For instance, FA chain lengths smaller than C24 readily formed self-assembled structures containing 2 (C16), 6 (C22, bis-C12), and 12 (C24) monomers, as measured by analytical ultracentrifugation (AUC). Incubation with albumin disrupted these supramolecular architectures to form FA-ASO/albumin complexes mostly with 2:1 stoichiometry and binding affinities in the low micromolar range, as determined by isothermal titration calorimetry (ITC) and analytical ultracentrifugation (AUC). Binding of FA-ASOs underwent a biphasic pattern for medium-length FA chain lengths (>C16) with an initial endothermic phase of particulate disruption, followed by an exothermic binding event to the albumin. Conversely, ASO modified with di-palmitic acid (C32) formed a strong, hexameric complex. This structure was not disrupted when incubated with albumin under conditions above the critical nanoparticle concentration (CNC; <0.4 µM). It is noteworthy that the interaction of parent, fatty acid-free malat1 ASO to albumin was below detectability by ITC (KD ≫150 µM). This work demonstrates that the nature of mono- vs multimeric structures of hydrophobically modified ASOs is governed by the hydrophobic effect. Consequently, supramolecular assembly to form particulate structures is a direct consequence of the fatty acid chain length. This provides opportunities to exploit the concept of hydrophobic modification to influence pharmacokinetics (PK) and biodistribution for ASOs in two ways: (1) binding of the FA-ASO to albumin as a carrier vehicle and (2) self-assembly resulting in albumin-inert, supramolecular architectures. Both concepts create opportunities to influence biodistribution, receptor interaction, uptake mechanism, and pharmacokinetics/pharmacodynamics (PK/PD) properties in vivo, potentially enabling access to extrahepatic tissues in sufficient concentration to treat disease.

DBA2J db/db mice are susceptible to early albuminuria and glomerulosclerosis that correlate with systemic insulin resistance.

Diabetic nephropathy (DN) is the leading cause of kidney failure in the world. To understand important mechanisms underlying this condition, and to develop new therapies, good animal models are required. In mouse models of type 1 diabetes, the DBA/2J strain has been shown to be more susceptible to develop kidney disease than other common strains. We hypothesized this would also be the case in type 2 diabetes. We studied db/db and wild-type (wt) DBA/2J mice and compared these with the db/db BLKS/J mouse, which is currently the most widely used type 2 DN model. Mice were analyzed from age 6 to 12 wk for systemic insulin resistance, albuminuria, and glomerular histopathological and ultrastructural changes. Body weight and nonfasted blood glucose were increased by 8 wk in both genders, while systemic insulin resistance commenced by 6 wk in female and 8 wk in male db/db DBA/2J mice. The urinary albumin-to-creatinine ratio (ACR) was closely linked to systemic insulin resistance in both sexes and was increased ~50-fold by 12 wk of age in the db/db DBA/2J cohort. Glomerulosclerosis, foot process effacement, and glomerular basement membrane thickening were observed at 12 wk of age in db/db DBA/2J mice. Compared with db/db BLKS/J mice, db/db DBA/2J mice had significantly increased levels of urinary ACR, but similar glomerular histopathological and ultrastructural changes. The db/db DBA/2J mouse is a robust model of early-stage albuminuric DN, and its levels of albuminuria correlate closely with systemic insulin resistance. This mouse model will be helpful in defining early mechanisms of DN and ultimately the development of novel therapies.

A framework for the in vitro evaluation of cancer-relevant molecular characteristics and mitogenic potency of insulin analogues.

Epidemiological and laboratory studies raise the possibility of a link between clinically prescribed insulin analogues and increased cancer risk. Accordingly, there is a regulatory mandate for cancer-related pre-clinical safety evaluation during insulin analogue development, but currently, there is no standardized framework for such in vitro evaluation. We tested human insulin; the super-mitogenic insulin, X10 and insulin-like growth factor I, in four cancer cell lines with a range of insulin-like growth factor-I receptor (IGF-IR)/IR (insulin receptor) ratios (HCT 116, HT-29, COLO 205 and MCF7) and related these to IGF-IR and IR expression in 17 human adenocarcinomas. All cell types were IR-A isoform dominant. We determined IGF-IR/IR signalling pathway endpoints in dose- and time-varying experiments, and performed mitogenic dose-response equivalent assays to derive EC50 values, and correlated these with IGF-IR/IR ratios. We superimposed relative EC50 values onto data from the literature in a meta-analysis. The IGF-IR/IR ratios varied from <1 to 12 in the selected cell lines; similar pattern ranges were observed in human adenocarcinomas. The three ligands demonstrated differential IR/IGF-IR and Akt phosphorylation, which correlated with cell-specific IGF-IR/IR ratios. Mitogenic profiles of X10 mimicked those for insulin-like growth factor I (IGF-I) and correlated with IGF-IR/IR ratios. The meta-analysis, adding data from five additional studies, supported the hypothesis that ligand mitogenic potency, relative to human insulin, increases with increasing cell-specific IGF-IR/IR ratio. This study established a framework for the in vitro evaluation of cancer-relevant bioassays for comparisons of insulin analogues, and specifically consolidated earlier studies that determination of the cell-specific IGF-IR/IR ratio is crucial for the interpretation of ranking relative biological activities.

Surface-expressed insulin receptors as well as IGF-I receptors both contribute to the mitogenic effects of human insulin and its analogues.

There is a medical need for new insulin analogues. Yet, molecular alterations to the insulin molecule can theoretically result in analogues with carcinogenic effects. Preclinical carcinogenicity risk assessment for insulin analogues rests to a large extent on mitogenicity assays in cell lines. We therefore optimized mitogenicity assay conditions for a panel of five cell lines. All cell lines expressed insulin receptors (IR), IGF-I receptors (IGF-IR) and hybrid receptors, and in all cell lines, insulin as well as the comparator compounds X10 and IGF-I caused phosphorylation of the IR as well as IGF-IR. Insulin exhibited mitogenicity EC(50) values in the single-digit nanomolar to picomolar range. We observed correlations across cell types between (i) mitogenic potency of insulin and IGF-IR/IR ratio, (ii) Akt phosphorylation and mitogenic potency and (iii) Akt phosphorylation and IR phosphorylation. Using siRNA-mediated knockdown of IR and IGF-IR, we observed that in HCT 116 cells the IR appeared dominant in driving the mitogenic response to insulin, whereas in MCF7 cells the IGF-IR appeared dominant in driving the mitogenic response to insulin. Together, our results show that the IR as well as IGF-IR may contribute to the mitogenic potency of insulin. While insulin was a more potent mitogen than IGF-I in cells expressing more IR than IGF-IR, the hyper-mitogenic insulin analogue X10 was a more potent mitogen than insulin across all cell types, supporting that the hyper-mitogenic effect of X10 involves the IR as well as the IGF-IR. These results are relevant for preclinical safety assessment of developmental insulin analogues.

Strontium is a biased agonist of the calcium-sensing receptor in rat medullary thyroid carcinoma 6-23 cells.

The calcium-sensing receptor (CaSR)-specific allosteric modulator cinacalcet has revolutionized the treatment of secondary hyperparathyroidism in patients with chronic kidney disease. However, its application is limited to patients with end-stage renal disease because of hypocalcemic side effects presumably caused by CaSR-mediated calcitonin secretion from thyroid parafollicular C-cells. These hypocalcemic side effects might be dampened by compounds that bias the signaling of CaSR, causing similar therapeutic effects as cinacalcet without stimulating calcitonin secretion. Because biased signaling of CaSR is poorly understood, the objective of the present study was to investigate biased signaling of CaSR by using rat medullary thyroid carcinoma 6-23 cells as a model of thyroid parafollicular C-cells. By doing concentration-response experiments we focused on the ability of two well known CaSR agonists, calcium and strontium, to activate six different signaling entities: G(q/11) signaling, G(i/o) signaling, G(s) signaling, extracellular signal-regulated kinases 1 and 2 (ERK1/2) signaling, intracellular calcium ([Ca(2+)](i)) mobilization, and calcitonin secretion. The experiments showed that strontium biases CaSR signaling toward ERK1/2 signaling and possibly another pathway independent of G(q/11) signaling and [Ca(2+)](i) mobilization. It is noteworthy that the potency of strontium-stimulated calcitonin secretion was elevated compared with calcium. Combining these results with experiments investigating signaling pathway components involved in calcitonin secretion, we found that the enhanced potency of strontium-mediated calcitonin secretion was caused by a different signaling pattern than that produced by calcium. Together, our results suggest that calcitonin secretion can be affected by CaSR-stimulated signaling bias, which may be used to develop novel drugs for the treatment of secondary hyperparathyroidism.

Efficient gene silencing by delivery of locked nucleic acid antisense oligonucleotides, unassisted by transfection reagents.

For the past 15-20 years, the intracellular delivery and silencing activity of oligodeoxynucleotides have been essentially completely dependent on the use of a delivery technology (e.g. lipofection). We have developed a method (called 'gymnosis') that does not require the use of any transfection reagent or any additives to serum whatsoever, but rather takes advantage of the normal growth properties of cells in tissue culture in order to promote productive oligonucleotide uptake. This robust method permits the sequence-specific silencing of multiple targets in a large number of cell types in tissue culture, both at the protein and mRNA level, at concentrations in the low micromolar range. Optimum results were obtained with locked nucleic acid (LNA) phosphorothioate gap-mers. By appropriate manipulation of oligonucleotide dosing, this silencing can be continuously maintained with little or no toxicity for >240 days. High levels of oligonucleotide in the cell nucleus are not a requirement for gene silencing, contrary to long accepted dogma. In addition, gymnotic delivery can efficiently deliver oligonucleotides to suspension cells that are known to be very difficult to transfect. Finally, the pattern of gene silencing of in vitro gymnotically delivered oligonucleotides correlates particularly well with in vivo silencing. The establishment of this link is of particular significance to those in the academic research and drug discovery and development communities.

Investigation of MicroRNA-134 as a Target against Seizures and SUDEP in a Mouse Model of Dravet Syndrome.

Dravet syndrome (DS) is a catastrophic form of pediatric epilepsy mainly caused by noninherited mutations in the SCN1A gene. DS patients suffer severe and life-threatening focal and generalized seizures which are often refractory to available anti-seizure medication. Antisense oligonucleotides (ASOs) based approaches may offer treatment opportunities in DS. MicroRNAs are short noncoding RNAs that play a key role in brain structure and function by post-transcriptionally regulating gene expression, including ion channels. Inhibiting miRNA-134 (miR-134) using an antimiR ASO (Ant-134) has been shown to reduce evoked seizures in juvenile and adult mice and reduce epilepsy development in models of focal epilepsy. The present study investigated the levels of miR-134 and whether Ant-134 could protect against hyperthermia-induced seizures, spontaneous seizures and mortality (SUDEP) in F1.Scn1a(+/-)tm1kea mice. At P17, animals were intracerebroventricular injected with 0.1-1 nmol of Ant-134 and subject to a hyperthermia challenge at postnatal day (P)18. A second cohort of P21 F1.Scn1a(+/-)tm1kea mice received Ant-134 and were followed by video and EEG monitoring until P28 to track the incidence of spontaneous seizures and SUDEP. Hippocampal and cortical levels of miR-134 were similar between wild-type (WT) and F1.Scn1a(+/-)tm1kea mice. Moreover, Ant-134 had no effect on hyperthermia-induced seizures, spontaneous seizures and SUDEP incidence were unchanged in Ant-134-treated DS mice. These findings suggest that targeting miR-134 does not have therapeutic applications in DS.

AntimiR targeting of microRNA-134 reduces seizures in a mouse model of Angelman syndrome.

Angelman syndrome (AS) is a severe neurodevelopmental disorder featuring ataxia, cognitive impairment, and drug-resistant epilepsy. AS is caused by mutations or deletion of the maternal copy of the paternally imprinted UBE3A gene, with current precision therapy approaches focusing on re-expression of UBE3A. Certain phenotypes, however, are difficult to rescue beyond early development. Notably, a cluster of microRNA binding sites was reported in the untranslated Ube3a1 transcript, including for miR-134, suggesting that AS may be associated with microRNA dysregulation. Here, we report levels of miR-134 and key targets are normal in the hippocampus of mice carrying a maternal deletion of Ube3a (Ube3a m-/p+ ). Nevertheless, intracerebroventricular injection of an antimiR oligonucleotide inhibitor of miR-134 (Ant-134) reduced audiogenic seizure severity over multiple trials in 21- and 42-day-old AS mice. Interestingly, Ant-134 also improved distance traveled and center crossings of AS mice in the open-field test. Finally, we show that silencing miR-134 can upregulate targets of miR-134 in neurons differentiated from Angelman patient-derived induced pluripotent stem cells. These findings indicate that silencing miR-134 and possibly other microRNAs could be useful to treat clinically relevant phenotypes with a later developmental window in AS.

Silencing of microRNA-155 in mice during acute inflammatory response leads to derepression of c/ebp Beta and down-regulation of G-CSF.

microRNA-155 (miR-155) has been implicated as a central regulator of the immune system, but its function during acute inflammatory responses is still poorly understood. Here we show that exposure of cultured macrophages and mice to lipopolysaccharide (LPS) leads to up-regulation of miR-155 and that the transcription factor c/ebp Beta is a direct target of miR-155. Interestingly, expression profiling of LPS-stimulated macrophages combined with overexpression and silencing of miR-155 in murine macrophages and human monocytic cells uncovered marked changes in the expression of granulocyte colony-stimulating factor (G-CSF), a central regulator of granulopoiesis during inflammatory responses. Consistent with these data, we show that silencing of miR-155 in LPS-treated mice by systemically administered LNA-antimiR results in derepression of the c/ebp Beta isoforms and down-regulation of G-CSF expression in mouse splenocytes. Finally, we report for the first time on miR-155 silencing in vivo in a mouse inflammation model, which underscores the potential of miR-155 antagonists in the development of novel therapeutics for treatment of chronic inflammatory diseases.

Antagomir-mediated suppression of microRNA-134 reduces kainic acid-induced seizures in immature mice.

MicroRNAs are short non-coding RNAs that negatively regulate protein levels and perform important roles in establishing and maintaining neuronal network function. Previous studies in adult rodents have detected upregulation of microRNA-134 after prolonged seizures (status epilepticus) and demonstrated that silencing microRNA-134 using antisense oligonucleotides, termed antagomirs, has potent and long-lasting seizure-suppressive effects. Here we investigated whether targeting microRNA-134 can reduce or delay acute seizures in the immature brain. Status epilepticus was induced in 21 day-old (P21) male mice by systemic injection of 5 mg/kg kainic acid. This triggered prolonged electrographic seizures and select bilateral neuronal death within the CA3 subfield of the hippocampus. Expression of microRNA-134 and functional loading to Argonaute-2 was not significantly changed in the hippocampus after seizures in the model. Nevertheless, when levels of microRNA-134 were reduced by prior intracerebroventricular injection of an antagomir, kainic acid-induced seizures were delayed and less severe and mice displayed reduced neuronal death in the hippocampus. These studies demonstrate targeting microRNA-134 may have therapeutic applications for the treatment of seizures in children.

Effects of obesity on insulin: insulin-like growth factor 1 hybrid receptor expression and Akt phosphorylation in conduit and resistance arteries.

Insulin and insulin-like growth factor-1 stimulate specific responses in arteries, which may be disrupted by diet-induced obesity. We examined (1) temporal effects of high-fat diet compared to low-fat diet in mice on insulin receptor, insulin-like growth factor-1 receptor, insulin receptor/insulin-like growth factor-1 receptor hybrid receptor expression and insulin/insulin-like growth factor-1-mediated Akt phosphorylation in aorta; and (2) effects of high-fat diet on insulin and insulin-like growth factor-1-mediated Akt phosphorylation and vascular tone in resistance arteries. Medium-term high-fat diet (5 weeks) decreased insulin-like growth factor-1 receptor expression and increased hybrid expression (~30%) only. After long-term (16 weeks) high-fat diet, insulin receptor expression was reduced by ~30%, insulin-like growth factor-1 receptor expression decreased a further ~40% and hybrid expression increased a further ~60%. Independent correlates of hybrid receptor expression were high-fat diet, duration of high-fat diet and plasma insulin-like growth factor-1 (all p < 0.05). In aorta, insulin was a more potent activator of Akt than insulin-like growth factor-1, whereas in resistance arteries, insulin-like growth factor-1 was more potent than insulin. High-fat diet blunted insulin-mediated vasorelaxation ( p < 0.01) but had no effect on insulin-like growth factor-1-mediated vasorelaxation in resistance arteries. Our findings support the possibility that hybrid receptor level is influenced by nutritional and metabolic cues. Moreover, vessel-dependent effects of insulin and insulin-like growth factor-1 on vascular tone and Akt activation may have implications in treating obesity-related vascular disease.

The function of BCL9 in Wnt/beta-catenin signaling and colorectal cancer cells.

BACKGROUND: Most cases of colorectal cancer are initiated by hyperactivation of the Wnt/beta-catenin pathway due to mutations in the APC tumour suppressor, or in beta-catenin itself. A recently discovered component of this pathway is Legless, which is essential for Wnt-induced transcription during Drosophila development. Limited functional information is available for its two mammalian relatives, BCL9 and B9L/BCL9-2: like Legless, these proteins bind to beta-catenin, and RNAi-mediated depletion of B9L/BCL9-2 has revealed that this protein is required for efficient beta-catenin-mediated transcription in mammalian cell lines. No loss-of-function data are available for BCL9. METHODS: We have used overexpression of dominant-negative forms of BCL9, and RNAi-mediated depletion, to study its function in human cell lines with elevated Wnt pathway activity, including colorectal cancer cells. RESULTS: We found that BCL9 is required for efficient beta-catenin-mediated transcription in Wnt-stimulated HEK 293 cells, and in the SW480 colorectal cancer cell line whose Wnt pathway is active due to APC mutation. Dominant-negative mutants of BCL9 indicated that its function depends not only on its beta-catenin ligand, but also on an unknown ligand of its C-terminus. Finally, we show that BCL9 and B9L are both Wnt-inducible genes, hyperexpressed in colorectal cancer cell lines, indicating that they are part of a positive feedback loop. CONCLUSION: BCL9 is required for efficient beta-catenin-mediated transcription in human cell lines whose Wnt pathway is active, including colorectal cancer cells, indicating its potential as a drug target in colorectal cancer.

Genetic and epigenetic alterations of the reduced folate carrier in untreated diffuse large B-cell lymphoma.

The reduced folate carrier (RFC) is a transmembrane protein that mediates cellular uptake of reduced folates and antifolate drugs, including methotrexate (MTX). Acquired alterations of the RFC gene have been associated with resistance to MTX in cancer cell lines and primary osteosarcomas. Here, we examined RFC for mutations and promoter hypermethylation in (i) the inherently MTX-resistant lymphoma cell line (RL); (ii) 30 paired cases of acute lymphoblastic leukemia (ALL) obtained at diagnosis and at relapse after treatment with MTX; and (iii) 25 cases of diffuse large B-cell lymphoma (DLBCL) at diagnosis, none of which had been previously exposed to MTX. Aberrant hypermethylation of the RFC promoter occurred in RL cells and two of the primary DLBCLs. In one additional DLBCL, a single-base substitution in RFC was identified, leading to the introduction of a premature termination codon (c.1396C>T; p.Q466X). A missense mutation affecting the 11th transmembrane domain of RFC (c.1250T>C; p.I417T) was found in one case of ALL at diagnosis. In ALL, RFC promoter hypermethylation was found neither at diagnosis nor at relapse and thus is not a common cause of low levels of RFC expression associated with adverse outcome. In DLBCL, genetic and epigenetic alterations of RFC were detected at diagnosis in the absence of a selective MTX pressure, suggesting that these alterations may possibly contribute to the development of lymphoma.

Genetic and epigenetic alterations of the APC gene in malignant melanoma.

High levels of beta-catenin and activating mutations in the beta-catenin gene (CTNNB1) have been demonstrated in malignant melanomas, implicating dysregulated Wnt signalling in the pathogenesis of this malignancy. We systematically examined melanoma cell lines for activating CTNNB1 mutations as well as genetic and epigenetic alterations of the adenomatous polyposis coli gene (APC), another key component of the Wnt signalling transduction pathway. Of 40 cell lines tested, one carried a truncating APC mutation and loss of the corresponding wild-type allele, and one carried a CTNNB1 missense mutation. Hypermethylation of APC promoter 1A was present in five of the cell lines (13%) and in nine of 54 melanoma biopsies (17%). Cells with truncating APC or activating CTNNB1 mutations showed increased transcription from endogenous and ectopic beta-catenin/T-cell factor (Tcf)-responsive target genes, consistent with the known effects of these alterations on beta-catenin stability and Tcf transactivation. In contrast, cell lines with APC promoter 1A hypermethylation did not show increased Wnt signalling, probably due to residual APC activity expressed from promoter 1B. Suppression of APC transcripts in melanoma cells by stable expression of short hairpin RNAs led to a Wnt signalling-independent increase in cell proliferation, but also reduced the invasive growth in collagen type I. Collectively, our data suggest that the tumour-suppressive function of APC in melanocytic cells is dose dependent. We propose that epigenetic silencing of promoter 1A may contribute to the development of malignant melanoma by reducing the expression of APC to a level that promotes cell proliferation without compromising the invasive capacity.

Distinct modes of deregulation of the proto-oncogenic Cdc25A phosphatase in human breast cancer cell lines.

The rapid cell cycle arrest in response to DNA damage is mediated by degradation of the Cdc25A phosphatase, a proto-oncogene whose mRNA is frequently overexpressed in human tumours. Here, we study the occurrence and mechanisms of Cdc25A deregulation in human breast cancer cell lines. We demonstrate aberrantly elevated Cdc25A protein abundance and phosphatase activity in eight out of 15 cell lines, in some cases resulting in abrogation of the Cdc25A-mediated checkpoint response to ionizing radiation (IR), and this defect correlated with hypersensitivity to IR. Furthermore, we present evidence that deregulation of Cdc25A occurs predominantly on the post-transcriptional level, as overabundant Cdc25A protein was usually not accompanied by adequate mRNA overexpression. Instead, we demonstrate that aberrantly enhanced protein stability is an important mechanism underlying Cdc25A overabundance in a subset of breast cancer cell lines. Given the frequency of this mechanism, we propose that the DNA integrity checkpoint controlling Cdc25A protein stability might be a common target for deregulation in breast cancer.

Frequent hypermethylation of DBC1 in malignant lymphoproliferative neoplasms.

Allelic loss at chromosome 9q31-34 is a frequent event in many lymphoproliferative malignancies. Here, we examined DBC1 at 9q33.1 as a potential target in lymphomagenesis. DBC1 is a putative tumor suppressor that has been shown to be involved in the regulation of cell growth and programmed cell death. The methylation status of the DBC1 promoter CpG island was examined by methylation-specific PCR, bisulfite sequencing, and methylation-specific melting curve analysis. DBC1 was hypermethylated in 5 of 5 B-cell-derived lymphoma cell lines, 41 of 42 diffuse large B-cell lymphomas, 24 of 24 follicular lymphomas, 5 of 5 mantle cell lymphomas, 4 of 4 small lymphocytic lymphomas, 1 of 2 lymphoplasmacytoid lymphomas, and in 12 of 12 acute lymphoblastic leukemias, but was unmethylated in 1 case of splenic marginal zone lymphoma, in 12 of 12 multiple myelomas, in 24 of 24 reactive lymph nodes, and in 12 of 12 samples of blood lymphocytes from random donors. DBC1 hypermethylation was associated with transcriptional silencing in lymphoma cell lines, and reexpression of this gene could be induced by treatment with the demethylating agent, 5-aza-2'-deoxycytidine. Our data suggest that hypermethylation of the DBC1 promoter region is a frequent event during the development of lymphoproliferative malignancies, and that DBC1 hypermethylation may serve as a marker for these cancers.

DNA methylation: an epigenetic pathway to cancer and a promising target for anticancer therapy.

The unique properties of a cancer cell are acquired through a stepwise accumulation of heritable changes in the information content of proto-oncogenes and tumor suppressor genes. While gain, loss, and mutation of genetic information have long been known to contribute to tumorigenesis, it has been increasingly recognized over the past 5 years that 'epigenetic' mechanisms may play an equally important role. The main epigenetic modification of the human genome is methylation of cytosine residues within the context of the CpG dinucleotide. De novo methylation of 'CpG islands' in the promoter regions of tumor suppressor genes may lead to transcriptional silencing through a complex process involving histone deacetylation and chromatin condensation, and thus represents a tumorigenic event that is functionally equivalent to genetic changes like mutation and deletion. DNA methylation is interesting from a diagnostic viewpoint because it may be easily detected in DNA released from neoplastic and preneoplastic lesions into serum, urine or sputum, and from a therapeutic viewpoint because epigenetically silenced genes may be reactivated by inhibitors of DNA methylation and/or histone deacetylase. A better understanding of epigenetic mechanisms leading to tumor formation and chemoresistance may eventually improve current cancer treatment regimens and be instructive for a more rational use of anticancer agents.

Profiling DNA methylation by melting analysis.

The idea of modifying DNA with bisulfite has paved the way for a variety of polymerase chain reaction (PCR) methods for accurately mapping 5-methylcytosine at specific genes. Bisulfite selectively deaminates cytosine to uracil under conditions where 5-methylcytosine remains unreacted. Following conventional PCR amplification of bisulfite-treated DNA, original cytosines appear as thymine while 5-methylcytosines appear as cytosine. Because the relative thermostability of a DNA duplex increases with increasing content of G:C base pairs, PCR products originating from DNA templates with different contents of 5-methylcytosine differ in melting temperature, i.e., the temperature required to convert the double helix into random coils. We describe two methods that resolve differentially methylated DNA sequences on the basis of differences in melting temperature. The first method integrates PCR amplification of bisulfite-treated DNA and subsequent melting analysis by using a thermal cycler coupled with a fluorometer. By including in the reaction a PCR-compatible, fluorescent dye that specifically binds to double-stranded DNA, the melting properties of the PCR product can be examined directly in the PCR tube by continuous fluorescence monitoring during a temperature transition. The second method relies on resolution of alleles with different 5-methylcytosine contents by analysis of PCR products in a polyacrylamide gel containing a gradient of chemical denaturants. Optimal resolution of differences in melting temperature is achieved by a special design of PCR primers. Both methods allow resolution of "heterogeneous" methylation, i.e., the situation where the content and distribution of 5-methylcytosine in a target gene differ between different molecules in the same sample.

ATM mutations are associated with inactivation of the ARF-TP53 tumor suppressor pathway in diffuse large B-cell lymphoma.

The ATM serine-threonine kinase plays a central role in the cellular response to DNA damage. Germ-line mutations in the ATM gene cause ataxia-telangiectasia (A-T), a multisystem disorder associated with predisposition to lymphoma and acute leukemia. Moreover, somatic ATM mutations have been identified in T-cell prolymphocytic leukemia, mantle cell lymphoma, and B-cell chronic lymphocytic leukemia. In this study, the entire ATM coding sequence was examined in genomic DNA from 120 lymphoid neoplasms. Novel mutations and mutations implicated in cancer and/or A-T were found in 9 of 45 diffuse large B-cell lymphomas (DLBCLs), 2 of 24 follicular lymphomas, and 1 of 27 adult acute lymphoblastic leukemias, whereas no such mutations were detected among 24 peripheral T-cell lymphomas. The mutational spectrum consisted of 2 nonsense mutations, 1 mutation affecting RNA splicing, and 10 missense variants. Most of these mutations were associated with loss or mutation of the paired ATM allele, consistent with biallelic inactivation of ATM. Of the 9 DLBCLs with ATM mutations, 7 also carried TP53 mutations and/or deletions of the INK4a/ARF locus (P =.003). The ATM 735C>T substitution previously considered a rare normal variant was found to be 5.6 times more frequent in individuals with DLBCL than in random individuals (P =.026), suggesting that it may predispose to B-cell lymphoma. Our data suggest that ATM mutations contribute to the development of DLBCL, and that ATM and the ARF-p53 tumor suppressor pathway may cooperate in the pathogenesis of this malignancy.