A randomised, phase II trial of the DNA-hypomethylating agent 5-aza-2'-deoxycytidine (decitabine) in combination with carboplatin vs carboplatin alone in patients with recurrent, partially platinum-sensitive ovarian cancer.

BACKGROUND: Our previous laboratory and clinical data suggested that one mechanism underlying the development of platinum resistance in ovarian cancer is the acquisition of DNA methylation. We therefore tested the hypothesis that the DNA hypomethylating agent 5-aza-2'-deoxycytodine (decitabine) can reverse resistance to carboplatin in women with relapsed ovarian cancer. METHODS: Patients progressing 6-12 months after previous platinum therapy were randomised to decitabine on day 1 and carboplatin (AUC 6) on day 8, every 28 days or carboplatin alone. The primary objective was response rate in patients with methylated hMLH1 tumour DNA in plasma. RESULTS: After a pre-defined interim analysis, the study closed due to lack of efficacy and poor treatment deliverability in 15 patients treated with the combination. Responses by GCIG criteria were 9 out of 14 vs 3 out of 15 and by RECIST were 6 out of 13 vs 1 out of 12 for carboplatin and carboplatin/decitabine, respectively. Grade 3/4 neutropenia was more common with the combination (60% vs 15.4%) as was G2/3 carboplatin hypersensitivity (47% vs 21%). CONCLUSIONS: With this schedule, the addition of decitabine appears to reduce rather than increase the efficacy of carboplatin in partially platinum-sensitive ovarian cancer and is difficult to deliver. Patient-selection strategies, different schedules and other demethylating agents should be considered in future combination studies.

Recombination. Pieces of the site-specific recombination puzzle.

Understanding how nucleoprotein complexes interact with specific DNA sequences has come closer with structural and mechanistic studies of the interaction between the recombination enzyme resolvase and DNA.

SfiI endonuclease activity is strongly influenced by the non-specific sequence in the middle of its recognition site.

The SfiI endonuclease cleaves DNA at the sequence GGCCNNNN NGGCC, where N is any base and downward arrow is the point of cleavage. Proteins that recognise discontinuous sequences in DNA can be affected by the unspecified sequence between the specified base pairs of the target site. To examine whether this applies to SFII, a series of DNA duplexes were made with identical sequences apart from discrete variations in the 5 bp spacer. The rates at which SFII cleaved each duplex were measured under steady-state conditions: the steady-state rates were determined by the DNA cleavage step in the reaction pathway. SFII cleaved some of these substrates at faster rates than other substrates. For example, the change in spacer sequence from AACAA to AAACA caused a 70-fold increase in reaction rate. In general, the extrapolated values for k(cat) and K(m) were both higher on substrates with inflexible spacers than those with flexible structures. The dinucleotide at the site of cleavage was largely immaterial. SFII activity is thus highly dependent on conformational variations in the spacer DNA.

Structural analysis of a mutational hot-spot in the EcoRV restriction endonuclease: a catalytic role for a main chain carbonyl group.

Following random mutagenesis of the Eco RV endonuclease, a high proportion of the null mutants carry substitutions at Gln69. Such mutants display reduced rates for the DNA cleavage step in the reaction pathway, yet the crystal structures of wild-type Eco RV fail to explain why Gln69 is crucial for activity. In this study, crystal structures were determined for two mutants of Eco RV, with Leu or Glu at residue 69, bound to specific DNA. The structures of the mutants are similar to the native protein and no function can be ascribed to the side chain of the amino acid at this locus. Instead, the structures of the mutant proteins suggest that the catalytic defect is due to the positioning of the main chain carbonyl group. In the enzyme-substrate complex for Eco RV, the main chain carbonyl of Gln69 makes no interactions with catalytic functions but, in the enzyme-product complex, it coordinates a metal ion bound to the newly liberated 5'-phosphate. This re-positioning may be hindered in the mutant proteins. Molecular dynamics calculations indicate that the metal on the phosphoryl oxygen interacts with the carbonyl group upon forming the pentavalent intermediate during phosphodiester hydrolysis. A main chain carbonyl may thus play a role in catalysis by Eco RV.

Chromosomal localization, genomic organization and evolution of the genes encoding human phosphatidylinositol transfer protein membrane-associated (PITPNM) 1, 2 and 3.

Eukaryotic proteins containing a phosphatidylinositol transfer (PITP) domain can be divided into two groups, one consisting of small soluble 35-kDa proteins and the other those that are membrane-associated and show sequence similarities to the Drosophila retinal degeneration B (rdgB) protein. The rdgB protein consists of four domains, an amino terminal PITP domain, a Ca2+-binding domain, a transmembrane domain and a carboxyl terminal domain that interacts with the protein tyrosine kinase PYK2. Three mammalian phosphatidylinositol transfer protein membrane-associated genes (PITPNM1, 2 and 3) with homology to Drosophila rdgB have previously been described and shown to be expressed in the mammalian retina. These findings and the demonstration that the rdgB gene plays a critical role in the invertebrate phototransduction pathway have led to the mammalian genes being considered as candidate genes for human eye diseases. In order to facilitate the analysis of these genes we have used radiation hybrid mapping and fluorescence in situ hybridization to localize the PITPNM2 and 3 genes to human chromosomes 12p24 and 17p13 respectively and hybrid mapping to confirm the localization of PITPNM1 to chromosome 11q13. We have also determined the genomic organization of both the soluble and membrane-associated Drosophila and human PITP domain-containing genes. Phylogenetic analysis indicates that the two groups arose by gene duplication that occurred very early in animal evolution.

DNA cleavage at two recognition sites by the SfiI restriction endonuclease: salt dependence of cis and trans interactions between distant DNA sites.

At low ionic strength, the SfiI restriction enzyme cleaved at similar rates both supercoiled and linear DNA with two SfiI sites and linear DNA with one SfiI site. For the substrates with two sites, the majority of the DNA was converted directly to products cut at both sites; the enzyme appears to bind to two sites before catalyzing its reactions, looping out the intervening DNA. At high ionic strength, linear DNA with one SfiI site was not cut at all, linear DNA with two sites was cleaved slowly while supercoiled DNA with two sites was cleaved rapidly, though only half of the DNA with two sites was cut at both sites; the DNA that had been cut at one site was not cleaved again at the remaining site. The singly cut product must therefore have been generated by a reaction incorporating both sites. All DNA cleavage reactions by SfiI thus involve the tetrameric enzyme bound to two copies of its recognition sequence, but weakened DNA-protein interactions at high ionic strength can cause this complex to dissociate before cleaving both sites. Intramolecular interactions between distant DNA sites are generally thought to be enhanced by supercoiling and to be more stable than intermolecular interactions. The preference of SfiI at high ionic strength for substrates with two sites over substrates with one site and, in the former case, for supercoiled over linear DNA, validates this view. At low ionic strength, the similar rates with the different substrates may be due to rate-limiting product dissociation.

Site-specific recombination at res sites containing DNA-binding sequences for both Tn21 resolvase and CAP.

The res sites, the loci for site-specific recombination by resolvase, contain three binding sites for the protein; the cross-over site, I, and accessory sites II and III. The role of DNA bending by resolvase was examined by replacing either II or III in the res site from Tn21 with the recognition sequence for a heterologous DNA-bending protein, CAP (the catabolite-gene activator protein from Escherichia coli). The CAP sequence was placed at either the same position as the target sequence for Tn21 resolvase or a different position along the DNA. The activity of Tn21 resolvase for recombination between each hybrid and a wild-type res site was measured in the presence of CAP and cyclic AMP. When III was substituted, CAP inhibited Tn21 recombination, except when the CAP sequence was placed sufficiently far away from site II to allow resolvase to bind non-specifically to the DNA between II and the CAP site. With the substitutions at II, the extent of Tn21 recombination in the presence of CAP varied with the position of the CAP sequence: more recombination was observed when it superimposed the target sequence for resolvase than when it was displaced by five base-pairs. Efficient recombination by Tn21 resolvase thus seems to demand the cognate protein at site III in res, presumably for protein-protein interactions in the synaptic complex, while the function of resolvase at site II can be fulfilled, at least in part, by a heterologous DNA bend.

Reactions of the eco RV restriction endonuclease with fluorescent oligodeoxynucleotides: identical equilibrium constants for binding to specific and non-specific DNA.

The EcoRV restriction endonuclease cleaves DNA specifically at its recognition sequence in the presence of magnesium ions, but several studies have indicated that it binds to DNA in the absence of Mg2+ without any preference for its recognition site. However, specific binding to the recognition site has also been reported. To distinguish between these reports, oligodeoxynucleotides were tagged with either dansyl or eosin fluorophores at their 5' termini and annealed to form duplexes of 12 to 16 base-pairs. For each length of duplex, one derivative had the EcoRV recognition sequence while another lacked this sequence. For the duplexes with the recognition site, the fluorophores had no effect on DNA cleavage rates by EcoRV in the presence of Mg2+. The binding of the specific and non-specific duplexes to EcoRV in the absence of Mg2+ was measured by fluorescence resonance energy transfer and by fluorescence depolarization. In both procedures, the signal from the specific complex differed from the complex with non-specific DNA, with the depolarization data indicating that non-specific DNA bound to EcoRV retains a higher rotational freedom than specific DNA. Even so, the equilibrium constant for the binding of specific DNA was identical, within error limits, to that for non-specific DNA.

DNA looping by the Sfi I restriction endonuclease.

The SfiI endonuclease has to interact with two copies of its recognition sequence before it can cleave DNA. To demonstrate that the reaction of SfiI on a DNA with two sites involves the formation of a DNA loop, and to characterise the looping interactions on supercoiled and linear DNA, a series of plasmids was constructed with lengths of DNA between two SfiI sites varying from 104 to 211 bp. Both supercoiled and linear forms of each DNA were tested as substrates for SfiI. The reactions were monitored from the rates of DNA cleavage and from the generation of partially cleaved products, the latter indicating loop disruption before cleavage of both sites. On both supercoiled and linear DNA, the stabilities of the complexes spanning two SfiI sites varied in sinusoidal fashion with the distance between the sites, in the manner characteristic of a process governed by the helical periodicity of DNA. In all cases, the looping interaction was stabilised by DNA supercoiling. The sinusoidal variation from SfiI reactions on supercoiled DNA at 50 degreesC yielded a helical repeat of about 11.5 base-pairs per turn.

Communications between distant sites on supercoiled DNA from non-exponential kinetics for DNA synapsis by resolvase.

To determine how distant sites on supercoiled DNA communicate with each other, the mechanism of site-specific recombination by resolvase was analysed by using a rapid-reaction quench-flow device to study the kinetics of individual steps in the reaction pathway. Three sets of measurements revealed the rates for: (1) the initial binding of the protein to its target sites on the DNA; (2) the synapsis of the two DNA-protein complexes; (3) the overall process of recombination. The binding of the protein to the DNA was complete within 50 milliseconds while recombination required 500 seconds. Surprisingly, synapsis spanned this entire time range: some DNA molecules gave synaptic complexes within ten milliseconds after the initial binding, while others took over 100 seconds. The departure from exponential behaviour may be due to each molecule of DNA having to undergo different conformational fluctuations in order to juxtapose the recombinational sites. From polymer physics theory, the rate of synapsis ought to vary with either the size of the DNA molecule or the length of DNA between the recombinational sites, depending on the nature of the fluctuations, but plasmids of different sizes and with different spacings between the sites all gave the same rates for synapsis. This observation cannot be reconciled with current models for encounters of distant sites on supercoiled DNA. However, the superhelical axis in the DNA molecules used here will be branched at one or more positions and the encounters may arise from the motion of a single branch relative to the remainder of the chain. Alternatively, the non-exponential kinetics for synapsis may be due to multiple re-arrangements of non-productive complexes following DNA juxtaposition.

Site-specific recombination at res sites containing DNA-binding sequences for both Tn21 and Tn3 resolvases.

Tn3 and gamma delta resolvases catalyse site-specific recombination at res sites from Tn3 but not at Tn21 res sites. Tn21 resolvase has no activity at Tn3 sites and acts only at Tn21 sites. In both Tn3 and Tn21, res had three binding sites for the cognate resolvases; the cross-over site, I; and the accessory sites II and III, from which the bound proteins may stabilize the synaptic complex by protein-protein interactions. In this study hybrid res sites were made by replacing either II or III in the Tn21 res site with the equivalent sequence from Tn3. Plasmids containing either a hybrid and a wild-type Tn21 res site, or two hybrid sites, were tested for recombination. Relative to the reaction with two wild-type sites, recombination by Tn21 resolvase was reduced by replacing II at one res site and it was reduced further by replacing II at both loci but, in both cases, Tn21 recombination was enhanced by Tn3 or gamma delta resolvases. Very few of the amino acid on the external surface of gamma delta resolvases are conserved in Tn21. Moreover, mutants of gamma delta resolvase with defective protein-protein interactions also enhanced Tn21 recombination at this hybrid site. The resolvase at II thus seems not to be involved in protein-protein interactions and its main role may be to bend the DNA to the required structure. The replacement of III in the Tn21 site with Tn3 sequence also reduced recombination by Tn21 resolvase, especially when both loci carried the alteration but, in contrast to before, Tn3 or gamma delta resolvases now inhibited the Tn21 reaction. Recombination thus seems to require identical proteins at I and III, perhaps to allow for protein-protein interactions.

DNA cleavage reactions by type II restriction enzymes that require two copies of their recognition sites.

Several type II restriction endonucleases interact with two copies of their target sequence before they cleave DNA. Three such enzymes, NgoMIV, Cfr10I and NaeI, were tested on plasmids with one or two copies of their recognition sites, and on catenanes containing two interlinked rings of DNA with one site in each ring. The enzymes showed distinct patterns of behaviour. NgoMIV and NaeI cleaved the plasmid with two sites faster than that with one site and the catenanes at an intermediate rate, while Cfr10I gave similar steady-state rates on all three substrates. Both Cfr10I and NgoMIV converted the majority of the substrates with two sites directly to the products cut at both sites, while NaeI cleaved just one site at a time. All three enzymes thus synapse two DNA sites through three-dimensional space before cleaving DNA. With Cfr10I and NgoMIV, both sites are cleaved in one turnover, in a manner consistent with their tetrameric structures, while the cleavage of a single site by NaeI indicates that the second site acts not as a substrate but as an activator, as reported previously. The complexes spanning two sites have longer lifetimes on catenanes with one site in each ring than on circular DNA with two sites, which indicates that the catenanes have more freedom for site juxtaposition than plasmids with sites in cis.

DNA binding by mutants of Tn21 resolvase with DNA recognition functions from Tn3 resolvase.

Substitution of amino acids within the section of Tn21 resolvase that corresponds to a helix-turn-helix structure, with the equivalent residues from Tn3 resolvase, yields proteins that retain the ability to mediate recombination between res sites from Tn21. These proteins had no recombinational activity on res sites from Tn3, even when the complete recognition helix had been exchanged. In this study, the binding of these mutants of Tn21 resolvase to DNA fragments containing res from either Tn21 or Tn3 was analysed by DNase I footprinting and by gel retardation. With DNA containing res from Tn21, the mutants bound to all three of the binding sites for resolvase (I, II, and III) but with a lower affinity than wild-type Tn21 resolvase. No complexes were detected between Tn3 resolvase and Tn21 DNA. With DNA containing res from Tn3, both the mutants and wild-type Tn21 resolvase bound to sites II and III, forming similar complexes to those with Tn3 resolvase: some of the mutants had higher affinities for these two sites on Tn3 DNA than on Tn21 DNA. In contrast, at site I in res from Tn3 (the location of the recombinational cross-over), the derivatives of Tn21 resolvase formed aberrant complexes whose structures differed radically from that with Tn3 resolvase. Alterations in the amino acid sequence of resolvase, within the helix-turn-helix region, therefore modulate the affinity of the protein for its target sequence in the DNA, but the specificity of resolvase for recombination at its cognate res sites is determined by the resultant organization of the DNA-protein complex.

Alternative geometries of DNA looping: an analysis using the SfiI endonuclease.

Many processes are governed by proteins that bind to separate sites in DNA and loop out the intervening DNA, but the geometries of the loops have seldom been determined. The SfiI endonuclease cleaves DNA after interacting with two recognition sites, and is a favourable system for the analysis of DNA looping. A gel-shift assay was used here to examine the binding of SfiI to a series of linear DNA molecules containing two SfiI sites separated by 109-170 base-pairs. The complexes in which SfiI trapped a loop by binding to two sites in the same DNA were separated from the complexes containing SfiI bound to separate DNA molecules. Step-wise changes in the inter-site spacing generated two forms of the looped complex with different electrophoretic mobilities. The yields of each looped complex and the complexes from intermolecular synapses all varied cyclically with the inter-site spacing, with similar periodicities ( approximately 10.5 base-pairs) but with different phases. One looped complex predominated whenever the DNA between the sites needed to be underwound in order to produce the correct helical orientation of the binding sites. The other looped complex predominated whenever the intervening DNA needed to be overwound. We conclude that the former has trapped a right-handed loop with a negative node and the latter a left-handed loop with a positive node.

The SfiI restriction endonuclease makes a four-strand DNA break at two copies of its recognition sequence.

The SfiI endonuclease cleaves DNA by a mechanism that differs from other restriction enzymes. While most restriction enzymes are dimeric proteins that interact with a single DNA site, SfiI exists in solution as a tetramer and it appears to interact with two copies of its recognition sequence before it can cleave DNA. Its primary reaction is then to cut both strands at both sites in a concerted process. The two sites can be on either the same or different DNA molecules, so SfiI provides a test system for long-range interactions on DNA. On either supercoiled or linear DNA with two sites separated by 1 kb, the bridging interaction between the sites is an intramolecular event: the majority of the DNA is converted directly to products cleaved at both sites, bypassing intermediates cut at one site. Sites on separate DNA molecules, or two sites on linear DNA several kb apart, engage in an intermolecular interaction prior to cleavage. The interaction between two DNA molecules with one site on each is impeded by supercoiling in both partners but is permitted when one partner is linear: it may require reptation of one DNA through another. SfiI reactions have marked similarities to some of the reactions catalysed by site-specific recombination enzymes.

DNA cleavage by the EcoRV restriction endonuclease: pH dependence and proton transfers in catalysis.

To characterise the pH dependence of phosphodiester hydrolysis by the EcoRV endonuclease in the presence of Mn2+, single turnover reactions on a 12 bp DNA substrate were examined by stopped-flow and quench-flow methods between pH 6.0 and 8.5. At each pH value, the apparent rate constants for phosphodiester hydrolysis increased hyperbolically with the concentration of MnCl2, thus allowing values to be determined for the intrinsic rate constant at saturation with Mn2+ and the equilibrium dissociation constant for Mn2+. The equilibrium constants showed no systematic variation across the pH range tested, while the rate constants increased steeply with increasing pH up to an asymptote above pH 7.5. At low pH conditions, the gradient of a plot of log (rate constant) against pH approached a value of 2. DNA cleavage by EcoRV thus requires the de-protonation of two acidic groups. To determine whether aspartate 36 is one of the groups, mutants of EcoRV were made with other amino acid residues at position 36. Glutamate caused a partial loss of activity, while all other replacements gave near-zero activities. In contrast to wild-type EcoRV, the mutant with glutamate required the de-protonation of only one acidic group for DNA cleavage. A mechanism for EcoRV is proposed in which the water molecule that hydrolyses the phosphodiester bond is de-protonated by two Bronsted bases, probably the ionised forms of aspartate 36 and glutamate 45.

Random walk models for DNA synapsis by resolvase.

During site-specific recombination by resolvase, the protein binds to two sites on a supercoiled DNA molecule and the loaded sites then interact with each other to form a synaptic complex. The kinetics of synapsis show non-exponential behaviour extending over five log units of time and are independent of the length of the DNA molecule and the length of DNA between the sites. In this study, numerical models were developed in order to account for how fluctuations in the structure of supercoiled DNA might lead to the juxtaposition of distant sites in a manner consistent with the experimental data on synapsis by resolvase. Models where the juxtaposition arises from fluctuations around branch points in the superhelix failed to match the data: they yielded non-exponential kinetics but only over two log units of time and they predicted longer synapsis times for both larger DNA molecules and larger inter-site spacings. In another model, one fraction of the juxtaposition events gives rise directly to the productive complex while the remaining fraction initially yields a non-productive complex: the latter molecules undergo no further fluctuations until the abortive synapse dissociates at the end of a delay period. This model again failed to match the experimental data. However, the inclusion of three sorts of non-productive complexes, each with a different delay constant, led to progress curves that concurred with the data. Schemes were also developed to account for the juxtaposition of three sites at a branch point in supercoiled DNA.

Selection of non-specific DNA cleavage sites by the type IC restriction endonuclease EcoR124I.

The Type IC restriction endonuclease EcoR124I binds specifically to its recognition sequence but subsequently translocates non-specific DNA past the complex in an ATP-dependent mechanism. The enzyme thus has the potential to cleave DNA at loci distant from the recognition site. We have scrutinised the link between translocation and cleavage on linear and circular DNA substrates. On linear DNA carrying two recognition sites, the majority of cleavages at loci distant from the recognition site occurred between the two sites, regardless of the inter-site distance or relative orientations. On circular DNA carrying one site, distant cleavages occurred throughout the DNA but an equivalent linear molecule underwent considerably fewer cleavages at distant loci. These results agree with published models for DNA tracking. However, on every molecule investigated, discrete cleavage sites were also observed within +/-250 bp of the recognition sites. The localised cleavages were not confined to particular DNA sequences and were independent of DNA topology. We propose a model to account for both distant and localised cleavage events. The conformation of the DNA loop extruded during tracking may result in two DNA segments being held in proximity to the restriction moiety on the protein, one close to the EcoR124I site and another distant from the site: cleavage may occur in either segment. Alternatively, the cutting of DNA close to recognition sites may be the result of multiple nicks being generated in the expanding loop before any extensive translocation.

Site-specific recombination by mutants of Tn21 resolvase with DNA recognition functions from Tn3 resolvase.

The resolvases from the transposons Tn3 and Tn21 are homologous proteins but they possess distinct specificities for the DNA sequence at their respective res sites. The DNA binding domain of resolvase contains an amino acid sequence that can be aligned with the helix-turn-helix motif of other DNA binding proteins. Mutations in the gene for Tn21 resolvase were made by replacing the section of DNA that codes for the helix-turn-helix with synthetic oligonucleotides. Each mutation substituted one amino acid in Tn21 resolvase with either the corresponding residue from Tn3 resolvase or a residue that lacks hydrogen bonding functions. The ability of these proteins to mediate recombination between res sites from either Tn21 or Tn3 was measured in vivo and in vitro. With one exception, where a glutamate residue had been replaced by leucine, the activity of these mutants was similar to that of wild-type Tn21 resolvase. A further mutation was made in which the complete recognition helix of Tn21 resolvase was replaced with that from Tn3 resolvase. This protein retained activity in recombining Tn21 res sites, though at a reduced level relative to wild-type; the reduction can be assigned entirely to weakened binding to this DNA. Neither this mutant nor any other derivative of Tn21 resolvase had any detectable activity for recombination between res sites from Tn3. The exchange of this section of amino acid sequence between the two resolvases is therefore insufficient to alter the DNA sequence specificity for recombination.

DNA cleavage by the EcoRV restriction endonuclease: roles of divalent metal ions in specificity and catalysis.

The roles of divalent metal ions in DNA cleavage by the EcoRV endonuclease were studied by using Co2+ or Mn2+ as substitutes for the natural cofactor Mg2+. In steady-state experiments with a 12 bp oligonucleotide substrate, Co2+ yielded a similar turnover rate to that with Mg2+, but Mn2+ gave a slower rate. Single turnovers of EcoRV on this substrate were analysed by stopped-flow and quench-flow methods, to determine the rates for the formation of the ternary enzyme-DNA-metal complex, the hydrolysis of the phosphodiester bonds and the dissociation of the cleaved DNA. With Co2+, all three steps had similar rates to those with Mg2+. In contrast, Mn2+ gave a faster rate for phosphodiester hydrolysis than either Mg2+ or Co2+, but a slower rate for product dissociation, thus accounting for its low turnover rate. Single turnovers on plasmids also yielded faster rates for substrate hydrolysis with Mn2+ compared to Mg2+ and Co2+. Since Mn2+ gave the most rapid rates for the hydrolytic step, despite being less electronegative than Co2+, the function of the metal ion at the active site of EcoRV cannot be just the polarisation of the scissile phosphate. Moreover, the minimal scheme for the Co2+-catalysed reaction requires two metal ions for DNA cleavage. The metal ions seem to be involved in the precise positioning of both the substrate and the water that acts as the attacking nucleophile and in activating that water molecule. A model is presented to account for how two metal ions might fulfil these functions.