ApoE Isoforms Inhibit Amyloid Aggregation of Proinflammatory Protein S100A9.

Increasing evidence suggests that the calcium-binding and proinflammatory protein S100A9 is an important player in neuroinflammation-mediated Alzheimer's disease (AD). The amyloid co-aggregation of S100A9 with amyloid-ß (Aß) is an important hallmark of this pathology. Apolipoprotein E (ApoE) is also known to be one of the important genetic risk factors of AD. ApoE primarily exists in three isoforms, ApoE2 (Cys112/Cys158), ApoE3 (Cys112/Arg158), and ApoE4 (Arg112/Arg158). Even though the difference lies in just two amino acid residues, ApoE isoforms produce differential effects on the neuroinflammation and activation of the microglial state in AD. Here, we aim to understand the effect of the ApoE isoforms on the amyloid aggregation of S100A9. We found that both ApoE3 and ApoE4 suppress the aggregation of S100A9 in a concentration-dependent manner, even at sub-stoichiometric ratios compared to S100A9. These interactions lead to a reduction in the quantity and length of S100A9 fibrils. The inhibitory effect is more pronounced if ApoE isoforms are added in the lipid-free state versus lipidated ApoE. We found that, upon prolonged incubation, S100A9 and ApoE form low molecular weight complexes with stochiometric ratios of 1:1 and 2:1, which remain stable under SDS-gel conditions. These complexes self-assemble also under the native conditions; however, their interactions are transient, as revealed by glutaraldehyde cross-linking experiments and molecular dynamics (MD) simulation. MD simulation demonstrated that the lipid-binding C-terminal domain of ApoE and the second EF-hand calcium-binding motif of S100A9 are involved in these interactions. We found that amyloids of S100A9 are cytotoxic to neuroblastoma cells, and the presence of either ApoE isoforms does not change the level of their cytotoxicity. A significant inhibitory effect produced by both ApoE isoforms on S100A9 amyloid aggregation can modulate the amyloid-neuroinflammatory cascade in AD.

Prion Protein Octarepeat Domain Forms Transient ß-Sheet Structures upon Residue-Specific Binding to Cu(II) and Zn(II) Ions.

Misfolding of the cellular prion protein (PrPC) is associated with the development of fatal neurodegenerative diseases called transmissible spongiform encephalopathies (TSEs). Metal ions appear to play a crucial role in PrPC misfolding. PrPC is a combined Cu(II) and Zn(II) metal-binding protein, where the main metal-binding site is located in the octarepeat (OR) region. Thus, the biological function of PrPC may involve the transport of divalent metal ions across membranes or buffering concentrations of divalent metal ions in the synaptic cleft. Recent studies have shown that an excess of Cu(II) ions can result in PrPC instability, oligomerization, and/or neuroinflammation. Here, we have used biophysical methods to characterize Cu(II) and Zn(II) binding to the isolated OR region of PrPC. Circular dichroism (CD) spectroscopy data suggest that the OR domain binds up to four Cu(II) ions or two Zn(II) ions. Binding of the first metal ion results in a structural transition from the polyproline II helix to the ß-turn structure, while the binding of additional metal ions induces the formation of ß-sheet structures. Fluorescence spectroscopy data indicate that the OR region can bind both Cu(II) and Zn(II) ions at neutral pH, but under acidic conditions, it binds only Cu(II) ions. Molecular dynamics simulations suggest that binding of either metal ion to the OR region results in the formation of ß-hairpin structures. As the formation of ß-sheet structures can be a first step toward amyloid formation, we propose that high concentrations of either Cu(II) or Zn(II) ions may have a pro-amyloid effect in TSE diseases.

The Binding of Different Substrate Molecules at the Docking Site and the Active Site of γ-Secretase Can Trigger Toxic Events in Sporadic and Familial Alzheimer's Disease.

Pathogenic changes in γ-secretase activity, along with its response to different drugs, can be affected by changes in the saturation of γ-secretase with its substrate. We analyze the saturation of γ-secretase with its substrate using multiscale molecular dynamics studies. We found that an increase in the saturation of γ-secretase with its substrate could result in the parallel binding of different substrate molecules at the docking site and the active site. The C-terminal domain of the substrate bound at the docking site can interact with the most dynamic presenilin sites at the cytosolic end of the active site tunnel. Such interactions can inhibit the ongoing catalytic activity and increase the production of the longer, more hydrophobic, and more toxic Aß proteins. Similar disruptions in dynamic presenilin structures can be observed with different drugs and disease-causing mutations. Both, C99-ßCTF-APP substrate and its different Aß products, can support the toxic aggregation. The aggregation depends on the substrate N-terminal domain. Thus, the C99-ßCTF-APP substrate and ß-secretase path can be more toxic than the C83-αCTF-APP substrate and α-secretase path. Nicastrin can control the toxic aggregation in the closed conformation. The binding of the C99-ßCTF-APP substrate to γ-secretase can be controlled by substrate channeling between the nicastrin and ß-secretase. We conclude that the presented two-substrate mechanism could explain the pathogenic changes in γ-secretase activity and Aß metabolism in different sporadic and familial cases of Alzheimer's disease. Future drug-development efforts should target different cellular mechanisms that regulate the optimal balance between γ-secretase activity and amyloid metabolism.

Co-Aggregation of S100A9 with DOPA and Cyclen-Based Compounds Manifested in Amyloid Fibril Thickening without Altering Rates of Self-Assembly.

The amyloid cascade is central for the neurodegeneration disease pathology, including Alzheimer's and Parkinson's, and remains the focus of much current research. S100A9 protein drives the amyloid-neuroinflammatory cascade in these diseases. DOPA and cyclen-based compounds were used as amyloid modifiers and inhibitors previously, and DOPA is also used as a precursor of dopamine in Parkinson's treatment. Here, by using fluorescence titration experiments we showed that five selected ligands: DOPA-D-H-DOPA, DOPA-H-H-DOPA, DOPA-D-H, DOPA-cyclen, and H-E-cyclen, bind to S100A9 with apparent Kd in the sub-micromolar range. Ligand docking and molecular dynamic simulation showed that all compounds bind to S100A9 in more than one binding site and with different ligand mobility and H-bonds involved in each site, which all together is consistent with the apparent binding determined in fluorescence experiments. By using amyloid kinetic analysis, monitored by thioflavin-T fluorescence, and AFM imaging, we found that S100A9 co-aggregation with these compounds does not hinder amyloid formation but leads to morphological changes in the amyloid fibrils, manifested in fibril thickening. Thicker fibrils were not observed upon fibrillation of S100A9 alone and may influence the amyloid tissue propagation and modulate S100A9 amyloid assembly as part of the amyloid-neuroinflammatory cascade in neurodegenerative diseases.

Polyoxometalates as Effective Nano-inhibitors of Amyloid Aggregation of Pro-inflammatory S100A9 Protein Involved in Neurodegenerative Diseases.

Pro-inflammatory and amyloidogenic S100A9 protein is central to the amyloid-neuroinflammatory cascade in neurodegenerative diseases. Polyoxometalates (POMs) constitute a diverse group of nanomaterials, which showed potency in amyloid inhibition. Here, we have demonstrated that two selected nanosized niobium POMs, Nb10 and TiNb9, can act as potent inhibitors of S100A9 amyloid assembly. Kinetics analysis based on ThT fluorescence experiments showed that addition of either Nb10 or TiNb9 reduces the S100A9 amyloid formation rate and amyloid quantity. Atomic force microscopy imaging demonstrated the complete absence of long S100A9 amyloid fibrils at increasing concentrations of either POM and the presence of only round-shaped and slightly elongated aggregates. Molecular dynamics simulation revealed that both Nb10 and TiNb9 bind to native S100A9 homo-dimer by forming ionic interactions with the positively charged Lys residue-rich patches on the protein surface. The acrylamide quenching of intrinsic fluorescence showed that POM binding does not perturb the Trp 88 environment. The far and near UV circular dichroism revealed no large-scale perturbation of S100A9 secondary and tertiary structures upon POM binding. These indicate that POM binding involves only local conformational changes in the binding sites. By using intrinsic and 8-anilino-1-naphthalene sulfonate fluorescence titration experiments, we found that POMs bind to S100A9 with a Kd of ca. 2.5 µM. We suggest that the region, including Lys 50 to Lys 54 and characterized by high amyloid propensity, could be the key sequences involved in S1009 amyloid self-assembly. The inhibition and complete hindering of S100A9 amyloid pathways may be used in the therapeutic applications targeting the amyloid-neuroinflammatory cascade in neurodegenerative diseases.

Natural Compound from Olive Oil Inhibits S100A9 Amyloid Formation and Cytotoxicity: Implications for Preventing Alzheimer's Disease.

Polyphenolic compounds in the Mediterranean diet have received increasing attention due to their protective properties in amyloid neurodegenerative and many other diseases. Here, we have demonstrated for the first time that polyphenol oleuropein aglycone (OleA), which is the most abundant compound in olive oil, has multiple potencies for the inhibition of amyloid self-assembly of pro-inflammatory protein S100A9 and the mitigation of the damaging effect of its amyloids on neuroblastoma SH-SY5Y cells. OleA directly interacts with both native and fibrillar S100A9 as shown by intrinsic fluorescence and molecular dynamic simulation. OleA prevents S100A9 amyloid oligomerization as shown using amyloid oligomer-specific antibodies and cross-ß-sheet formation detected by circular dichroism. It decreases the length of amyloid fibrils measured by atomic force microscopy (AFM) as well as reduces the effective rate of amyloid growth and the overall amyloid load as derived from the kinetic analysis of amyloid formation. OleA disintegrates already preformed fibrils of S100A9, converting them into nonfibrillar and nontoxic aggregates as revealed by amyloid thioflavin-T dye binding, AFM, and cytotoxicity assays. At the cellular level, OleA targets S100A9 amyloids already at the membranes as shown by immunofluorescence and fluorescence resonance energy transfer, significantly reducing the amyloid accumulation in GM1 ganglioside containing membrane rafts. OleA increases overall cell viability when neuroblastoma cells are subjected to the amyloid load and alleviates amyloid-induced intracellular rise of reactive oxidative species and free Ca2+. Since S100A9 is both a pro-inflammatory and amyloidogenic protein, OleA may effectively mitigate the pathological consequences of the S100A9-dependent amyloid-neuroinflammatory cascade as well as provide protection from neurodegeneration, if used within the Mediterranean diet as a potential preventive measure.

Structural Analysis of the Simultaneous Activation and Inhibition of γ-Secretase Activity in the Development of Drugs for Alzheimer's Disease.

Significance: The majority of the drugs which target membrane-embedded protease γ-secretase show an unusual biphasic activation-inhibition dose-response in cells, model animals, and humans. Semagacestat and avagacestat are two biphasic drugs that can facilitate cognitive decline in patients with Alzheimer's disease. Initial mechanistic studies showed that the biphasic drugs, and pathogenic mutations, can produce the same type of changes in γ-secretase activity. Results: DAPT, semagacestat LY-411,575, and avagacestat are four drugs that show different binding constants, and a biphasic activation-inhibition dose-response for amyloid-ß-40 products in SH-SY5 cells. Multiscale molecular dynamics studies have shown that all four drugs bind to the most mobile parts in the presenilin structure, at different ends of the 29 Å long active site tunnel. The biphasic dose-response assays are a result of the modulation of γ-secretase activity by the concurrent binding of multiple drug molecules at each end of the active site tunnel. The drugs activate γ-secretase by facilitating the opening of the active site tunnel, when the rate-limiting step is the tunnel opening, and the formation of the enzyme-substrate complex. The drugs inhibit γ-secretase as uncompetitive inhibitors by binding next to the substrate, to dynamic enzyme structures which regulate processive catalysis. The drugs can modulate the production of different amyloid-ß catalytic intermediates by penetration into the active site tunnel, to different depths, with different flexibility and different binding affinity. Conclusions: Biphasic drugs and pathogenic mutations can affect the same dynamic protein structures that control processive catalysis. Successful drug-design strategies must incorporate transient changes in the γ-secretase structure in the development of specific modulators of its catalytic activity.

Substrate Channeling via a Transient Protein-Protein Complex: The case of D-Glyceraldehyde-3-Phosphate Dehydrogenase and L-Lactate Dehydrogenase.

Substrate channeling studies have frequently failed to provide conclusive results due to poor understanding of this subtle phenomenon. We analyzed the mechanism of NADH-channeling from D-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) to L-lactate Dehydrogenase (LDH) using enzymes from different cells. Enzyme kinetics studies showed that LDH activity with free NADH and GAPDH-NADH complex always take place in parallel. The channeling is observed only in assays that mimic cytosolic conditions where free NADH concentration is negligible and the GAPDH-NADH complex is dominant. Molecular dynamics and protein-protein interaction studies showed that LDH and GAPDH can form a leaky channeling complex only at the limiting NADH concentrations. Surface calculations showed that positive electric field between the NAD(H) binding sites on LDH and GAPDH tetramers can merge in the LDH-GAPDH complex. NAD(H)-channeling within the LDH-GAPDH complex can be an extension of NAD(H)-channeling within each tetramer. In the case of a transient LDH-(GAPDH-NADH) complex, the relative contribution from the channeled and the diffusive paths depends on the overlap between the off-rates for the LDH-(GAPDH-NADH) complex and the GAPDH-NADH complex. Molecular evolution or metabolic engineering protocols can exploit substrate channeling for metabolic flux control by fine-tuning substrate-binding affinity for the key enzymes in the competing reaction paths.

In silico design of the first DNA-independent mechanism-based inhibitor of mammalian DNA methyltransferase Dnmt1.

BACKGROUND: We use our earlier experimental studies of the catalytic mechanism of DNA methyltransferases to prepare in silico a family of novel mechanism-based inhibitors of human Dnmt1. Highly specific inhibitors of DNA methylation can be used for analysis of human epigenome and for the creation of iPS cells. RESULTS: We describe a set of adenosyl-1-methyl-pyrimidin-2-one derivatives as novel mechanism-based inhibitors of mammalian DNA methyltransferase Dnmt1. The inhibitors have been designed to bind simultaneously in the active site and the cofactor site and thus act as transition-state analogues. Molecular dynamics studies showed that the lead compound can form between 6 to 9 binding interactions with Dnmt1. QM/MM analysis showed that the upon binding to Dnmt1 the inhibitor can form a covalent adduct with active site Cys1226 and thus act as a mechanism-based suicide-inhibitor. The inhibitor can target DNA-bond and DNA-free form of Dnmt1, however the suicide-inhibition step is more likely to happen when DNA is bound to Dnmt1. The validity of presented analysis is described in detail using 69 modifications in the lead compound structure. In total 18 of the presented 69 modifications can be used to prepare a family of highly specific inhibitors that can differentiate even between closely related enzymes such as Dnmt1 and Dnmt3a DNA methyltransferases. CONCLUSIONS: Presented results can be used for preparation of some highly specific and potent inhibitors of mammalian DNA methylation with specific pharmacological properties.

Decrease in catalytic capacity of γ-secretase can facilitate pathogenesis in sporadic and Familial Alzheimer's disease.

BACKGROUND: Alzheimer's disease can be a result of an age-induced disparity between increase in cellular metabolism of Aß peptides and decrease in maximal activity of a membrane-embedded protease γ-secretase. RESULTS: We compared activity of WT γ-secretase with the activity of 6 FAD mutants in its presenilin-1 component and 5 FAD mutants in Aß-part of its APP substrate (Familial Alzheimer's disease). All 11 FAD mutations show linear correlation between the decrease in maximal activity and the clinically observed age-of-onset and age-of-death. Biphasic-inhibitors showed that a higher ratio between physiological Aß-production and the maximal activity of γ-secretase can be observed in cells that can facilitate pathogenic changes in Aß-products. For example, Aß production in cells with WT γ-secretase is at 11% of its maximal activity, with delta-exon-9 mutant at 26%, while with M139V mutant is at 28% of the maximal activity. In the same conditions, G384A mutant is fully saturated and at its maximal activity. Similarly, Aß production in cells with γ-secretase complex carrying Aph1AL component is 12% of its maximal activity, while in cells with Aph1B complex is 26% of its maximal activity. Similar to the cell-based studies, clinical studies of biphasic dose-response in plasma samples of 54 healthy individuals showed variable ratios between physiological Aß production and the maximal activity of γ-secretase. CONCLUSIONS: The increase in the ratio between physiological Aß production and maximal activity of γ-secretase can be an early sign of pathogenic processes in enzyme-based, cell-based, and clinical studies of sporadic and Familiar Alzheimer's disease.

Modulators of γ-secretase activity can facilitate the toxic side-effects and pathogenesis of Alzheimer's disease.

BACKGROUND: Selective modulation of different Aß products of an intramembrane protease γ-secretase, could be the most promising strategy for development of effective therapies for Alzheimer's disease. We describe how different drug-candidates can modulate γ-secretase activity in cells, by studying how DAPT affects changes in γ-secretase activity caused by gradual increase in Aß metabolism. RESULTS: Aß 1-40 secretion in the presence of DAPT shows biphasic activation-inhibition dose-response curves. The biphasic mechanism is a result of modulation of γ-secretase activity by multiple substrate and inhibitor molecules that can bind to the enzyme simultaneously. The activation is due to an increase in γ-secretase's kinetic affinity for its substrate, which can make the enzyme increasingly more saturated with otherwise sub-saturating substrate. The noncompetitive inhibition that prevails at the saturating substrate can decrease the maximal activity. The synergistic activation-inhibition effects can drastically reduce γ-secretase's capacity to process its physiological substrates. This reduction makes the biphasic inhibitors exceptionally prone to the toxic side-effects and potentially pathogenic. Without the modulation, γ-secretase activity on it physiological substrate in cells is only 14% of its maximal activity, and far below the saturation. SIGNIFICANCE: Presented mechanism can explain why moderate inhibition of γ-secretase cannot lead to effective therapies, the pharmacodynamics of Aß-rebound phenomenon, and recent failures of the major drug-candidates such as semagacestat. Novel improved drug-candidates can be prepared from competitive inhibitors that can bind to different sites on γ-secretase simultaneously. Our quantitative analysis of the catalytic capacity can facilitate the future studies of the therapeutic potential of γ-secretase and the pathogenic changes in Aß metabolism.

Modulation of γ-secretase activity by multiple enzyme-substrate interactions: implications in pathogenesis of Alzheimer's disease.

BACKGROUND: We describe molecular processes that can facilitate pathogenesis of Alzheimer's disease (AD) by analyzing the catalytic cycle of a membrane-imbedded protease γ-secretase, from the initial interaction with its C99 substrate to the final release of toxic Aß peptides. RESULTS: The C-terminal AICD fragment is cleaved first in a pre-steady-state burst. The lowest Aß42/Aß40 ratio is observed in pre-steady-state when Aß40 is the dominant product. Aß42 is produced after Aß40, and therefore Aß42 is not a precursor for Aß40. The longer more hydrophobic Aß products gradually accumulate with multiple catalytic turnovers as a result of interrupted catalytic cycles. Saturation of γ-secretase with its C99 substrate leads to 30% decrease in Aß40 with concomitant increase in the longer Aß products and Aß42/Aß40 ratio. To different degree the same changes in Aß products can be observed with two mutations that lead to an early onset of AD, ΔE9 and G384A. Four different lines of evidence show that γ-secretase can bind and cleave multiple substrate molecules in one catalytic turnover. Consequently depending on its concentration, NotchΔE substrate can activate or inhibit γ-secretase activity on C99 substrate. Multiple C99 molecules bound to γ-secretase can affect processive cleavages of the nascent Aß catalytic intermediates and facilitate their premature release as the toxic membrane-imbedded Aß-bundles. CONCLUSIONS: Gradual saturation of γ-secretase with its substrate can be the pathogenic process in different alleged causes of AD. Thus, competitive inhibitors of γ-secretase offer the best chance for a successful therapy, while the noncompetitive inhibitors could even facilitate development of the disease by inducing enzyme saturation at otherwise sub-saturating substrate. Membrane-imbedded Aß-bundles generated by γ-secretase could be neurotoxic and thus crucial for our understanding of the amyloid hypothesis and AD pathogenesis.

Dnmt1 structure and function.

Dnmt1, the principal DNA methyltransferase in mammalian cells, is a large and a highly dynamic enzyme with multiple regulatory features that can control DNA methylation in cells. This chapter highlights how insights into Dnmt1 structure and function can advance our understanding of DNA methylation in cells. The allosteric site(s) on Dnmt1 can regulate processes of de novo and maintenance DNA methylation in cells. Remaining open questions include which molecules, by what mechanism, bind at the allosteric site(s) in cells? Different phosphorylation sites on Dnmt1 can change its activity or ability to bind DNA target sites. Thirty-one different molecules are currently known to have physical and/or functional interaction with Dnmt1 in cells. The Dnmt1 structure and enzymatic mechanism offer unique insights into those interactions. The interacting molecules are involved in chromatin organization, DNA repair, cell cycle regulation, and apoptosis and also include RNA polymerase II, some RNA-binding proteins, and some specific Dnmt1-inhibitory RNA molecules. Combined insights from studies of different enzymatic features of Dnmt1 offer novel ideas for development of drug candidates, and can be used in selection of promising drug candidates from more than 15 different compounds that have been identified as possible inhibitors of DNA methylation in cells.

Mammalian cytosine DNA methyltransferase Dnmt1: enzymatic mechanism, novel mechanism-based inhibitors, and RNA-directed DNA methylation.

This is a review of the enzymatic mechanism of DNA methyltransferase Dnmt1 and analysis of its implications on regulation of DNA methylation in mammalian cells and design of novel mechanism-based inhibitors. The methylation reaction by Dnmt1 has different phases that depend on DNA substrate and allosteric regulation. Consequently, depending on the phase, the differences in catalytic rates between unmethylated and pre-methylated DNA can vary between 30-40 fold, 3-6 fold or only 1 fold. The allosteric site and the active site can bind different molecules. Allosteric activity depends on DNA sequence, methylation pattern and DNA structure (single stranded vs. double stranded). Dnmt1 binds poly(ADP-ribose) and some RNA molecules. The results on kinetic preferences, allosteric activity and binding preference of Dnmt1 are combined together in one comprehensive model mechanism that can address regulation of DNA methylation in cells; namely, inhibition of DNA methylation by poly(ADP-ribose), RNA-directed DNA methylation by methylated and unmethylated non-coding RNA molecules, and transient interactions between Dnmt1 and genomic DNA. Analysis of reaction intermediates showed that equilibrium between base-flipping and base-restacking events can be the key mechanism in control of enzymatic activity. The two events have equal but opposite effect on accumulation of early reaction intermediates and methylation rates. The accumulation of early reaction intermediates can be exploited to improve the current inhibitors of Dnmt1 and achieve inhibition without toxic modifications in genomic DNA. [1,2-dihydropyrimidin-2-one]-5-methylene-(methylsulfonium)-adenosyl is described as the lead compound.

Interaction between mammalian glyceraldehyde-3-phosphate dehydrogenase and L-lactate dehydrogenase from heart and muscle.

The exceptionally high protein concentration in living cells can favor functional protein-protein interactions that can be difficult to detect with purified proteins. In this study we describe specific interactions between mammalian D-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and L-lactate dehydrogenase (LDH) isozymes from heart and muscle. We use poly(ethylene-glycol) (PEG)-induced coprecipitation and native agarose electrophoresis as two independent methods uniquely suited to mimic some of the conditions that can favor protein-protein interaction in living cells. We found that GAPDH interacts with heart or muscle isozymes of LDH with approximately one-to-one stoichiometry. The interaction is specific; GAPDH shows interaction with two LDH isozymes that have very different net charge and solubility in PEG solution, while no interaction is observed with GAPDH from other species, other NAD(H) dehydrogenases, or other proteins that have very similar net charge and molecular mass. Analytical ultracentrifugation showed that the LDH and GAPDH complex is insoluble in PEG solution. The interaction is abolished by saturation with NADH, but not by saturation with NAD(+) in correlation with GAPDH solubility in PEG solution. The crystal structures show that GAPDH and LDH isozymes share complementary size, shape, and electric potential surrounding the active sites. The presented results suggest that GAPDH and LDH have a functional interaction that can affect NAD(+)/NADH metabolism and glycolysis in living cells.

Mechanism of allosteric regulation of Dnmt1's processivity.

We have analyzed the relationship between the allosteric regulation and processive catalysis of DNA methyltransferase 1 (Dnmt1). Processivity is described quantitatively in terms of turnover rate, DNA dissociation rate, and processivity probability. Our results provide further evidence that the active site and the allosteric sites on Dnmt1 can bind DNA independently. Dnmt1's processive catalysis on unmethylated DNA is partially inhibited when the allosteric site binds unmethylated DNA and fully inhibited when the allosteric site binds a single-stranded oligonucleotide inhibitor. The partial inhibition by unmethylated DNA is caused by a decrease in the turnover rate and an increase in the substrate DNA dissociation rate. Processive catalysis with premethylated DNA is not affected if the allosteric site is exposed to premethylated DNA but is fully inhibited if the allosteric site binds unmethylated DNA or poly(dA-dT). In sum, the occupancy of the allosteric site modulates the enzyme's commitment to catalysis, which reflects the nature of the substrate and the DNA bound at the allosteric site. Our in vitro results are consistent with the possibility that the processive action of Dnmt1 may be regulated in vivo by specific regulatory nucleic acids such as DNA, RNA, or poly(ADP-ribose).

Accommodation and repair of a UV photoproduct in DNA at different rotational settings on the nucleosome surface.

Cyclobutane-thymine dimers (CTDs), the most common DNA lesion induced by UV radiation, cause 30 degrees bending and 9 degrees unwinding of the DNA helix. We prepared site-specific CTDs within a short sequence bracketed by strong nucleosome-positioning sequences. The rotational setting of CTDs over one turn of the helix near the dyad center on the histone surface was analyzed by hydroxyl radical footprinting. Surprisingly, the position of CTDs over one turn of the helix does not affect the rotational setting of DNA on the nucleosome surface. Gel-shift analysis indicates that one CTD destabilizes histone-DNA interactions by 0.6 or 1.1 kJ/mol when facing away or toward the histone surface, respectively. Thus, 0.5 kJ/mol energy penalty for a buried CTD is not enough to change the rotational setting of sequences with strong rotational preference. The effect of rotational setting on CTD removal by nucleotide excision repair (NER) was examined using Xenopus oocyte nuclear extracts. The NER rates are only 2-3 times lower in nucleosomes and change by only 1.5-fold when CTDs face away or toward the histone surface. Therefore, in Xenopus nuclear extracts, the rotational orientation of CTDs on nucleosomes has surprisingly little effect on rates of repair. These results indicate that nucleosome dynamics and/or chromatin remodeling may facilitate NER in gaining access to DNA damage in nucleosomes.

DNA cytosine C5 methyltransferase Dnmt1: catalysis-dependent release of allosteric inhibition.

We followed the cytosine C(5) exchange reaction with Dnmt1 to characterize its preference for different DNA substrates, its allosteric regulation, and to provide a basis for comparison with the bacterial enzymes. We determined that the methyl transfer is rate-limiting, and steps up to and including the cysteine-cytosine covalent intermediate are in rapid equilibrium. Changes in these rapid equilibrium steps account for many of the previously described features of Dnmt1 catalysis and specificity including faster reactions with premethylated DNA versus unmethylated DNA, faster reactions with DNA in which guanine is replaced with inosine [poly(dC-dG) vs poly(dI-dC)], and 10-100-fold slower catalytic rates with Dnmt1 relative to the bacterial enzyme M.HhaI. Dnmt1 interactions with the guanine within the CpG recognition site can prevent the premature release of the target base and solvent access to the active site that could lead to mutagenic deamination. Our results suggest that the beta-elimination step following methyl transfer is not mediated by free solvent. Dnmt1 shows a kinetic lag in product formation and allosteric inhibition with unmethylated DNA that is not observed with premethylated DNA. Thus, we suggest the enzyme undergoes a slow relief from allosteric inhibition upon initiation of catalysis on unmethylated DNA. Notably, this relief from allosteric inhibition is not caused by self-activation through the initial methylation reaction, as the same effect is observed during the cytosine C(5) exchange reaction in the absence of AdoMet. We describe limitations in the Michaelis-Menten kinetic analysis of Dnmt1 and suggest alternative approaches.

The mechanism of target base attack in DNA cytosine carbon 5 methylation.

We measured the tritium exchange reaction on cytosine C(5) in the presence of AdoMet analogues to investigate the catalytic mechanism of the bacterial DNA cytosine methyltransferase M.HhaI. Poly(dG-dC) and poly(dI-dC) substrates were used to investigate the function of the active site loop (residues 80-99), stability of the extrahelical base, base flipping mechanism, and processivity on DNA substrates. On the basis of several experimental approaches, we show that methyl transfer is the rate-limiting pre-steady-state step. Further, we show that the active site loop opening contributes to the rate-limiting step during multiple cycles of catalysis. Target base activation and nucleophilic attack by cysteine 81 are fast and readily reversible. Thus, the reaction intermediates involving the activated target base and the extrahelical base are in equilibrium and accumulate prior to the slow methyl transfer step. The stability of the activated target base depends on the active site loop closure, which is dependent on the hydrogen bond between isoleucine 86 and the guanine 5' to the target cytosine. These interactions prevent the premature release of the extrahelical base and uncontrolled solvent access; the latter modulates the exchange reaction and, by implication, the mutagenic deamination reaction. The processive catalysis by M.HhaI is also regulated by the interaction between isoleucine 86 and the DNA substrate. Nucleophilic attack by cysteine 81 is partially rate limiting when the target base is not fully stabilized in the extrahelical position, as observed during the reaction with the Gln(237)Trp mutant or in the cytosine C(5) exchange reaction in the absence of the cofactor.