Charge hopping in DNA.

The efficiency of charge migration through stacked Watson-Crick base pairs is analyzed for coherent hole motion interrupted by localization on guanine (G) bases. Our analysis rests on recent experiments, which demonstrate the competition of hole hopping transitions between nearest neighbor G bases and a chemical reaction of the cation G(+) with water. In addition, it has been assumed that the presence of units with several adjacent stacked G bases on the same strand leads to the additional vibronic relaxation process (G(+)G...G) --> (GG...G)(+). The latter may also compete with the hole transfer from (G(+)G...G) to a single G site, depending on the relative positions of energy levels for G(+) and (G(+)G...G). A hopping model is proposed to take the competition of these three rate steps into account. It is shown that the model includes two important limits. One corresponds to the situation where the charge relaxation inside a multiple guanine unit is faster than hopping. In this case hopping is terminated by several adjacent G bases located on the same strand, as has been observed for the GGG triple. In the opposite, slow relaxation limit the GG...G unit allows a hole to migrate further in accord with experiments on strand cleavage exploiting GG pairs. We demonstrate that for base pair sequences with only the GGG triple, the fast relaxation limit of our model yields practically the same sequence- and distance dependencies as measurements, without invoking adjustable parameters. For sequences with a certain number of repeating adenine:thymine pairs between neighboring G bases, our analysis predicts that the hole transfer efficiency varies in inverse proportion to the sequence length for short sequences, with change to slow exponential decay for longer sequences. Calculations performed within the slow relaxation limit enable us to specify parameters that provide a reasonable fit of our numerical results to the hole migration efficiency deduced from experiments with sequences containing GG pairs. The relation of the results obtained to other theoretical and experimental studies of charge transfer in DNA is discussed. We propose experiments to gain a deeper insight into complicated kinetics of charge-transfer hopping in DNA.

DNA splicing by directed ligation (SDL).

Splicing by directed ligation (SDL) is a method of in-phase joining of PCR-generated DNA fragments that is based on a pre-designed combination of class IIS restriction endonuclease recognition and cleavage sites. Since these enzymes cleave outside of their recognition sites, the resulting sticky end can have any desired sequence, and the site itself can be removed and does not appear in the final spliced DNA product. SDL is based on the addition of class IIS recognition sites onto primers used to amplify DNA sequences. Cleavage of the PCR products results in elimination of the recognition site-containing flanking sequences and leaves the DNA fragments crowned with protruding ends. With careful design of the sticky ends, several segments can be ligated together in a predetermined order in a single reaction. SDL requires fewer rounds of amplification than overlap extension methods, and is particularly useful for creating a series of recombinants that differ in one segment.

Probing accessible sites for ribozymes on human acetylcholinesterase RNA.

In order to design ribozymes for the efficient cleavage of a human acetylcholinesterase (AChE) in vitro transcript, a completely randomized decadeoxyribonucleotide (dN10) was used in conjunction with RNase H to identify suitable sites for annealing. Based on the observed cleavage pattern, ribozymes were designed to cleave the transcript at these positions. Five ribozymes so designed proved to be efficient in the transcript cleavage (k(react)/Km ranged from 0.9 x 10(4) to 68.2 x 10(4) M(-1) min(-10)). The best was 150-fold more active than the best designed on the basis of the MFold program. Thus, the RNase H mapping demonstrated a high predictive power for favorable ribozyme cleavage sites. The digestion pattern with RNase H differed dramatically from that observed with the single-strand-specific mung bean nuclease.

A high-level prokaryotic expression system: synthesis of human interleukin 1 alpha and its receptor antagonist.

Synthetic intronless genes, coding for human interleukin 1 alpha (IL 1 alpha) and interleukin 1 receptor antagonist (IL1ra), have been expressed efficiently in a specially designed prokaryotic vector, pGMCE (a pGEM1 derivative), where the target gene forms the second part of a two-cistron system. The first part of the system is a translation enhancer-containing mini-cistron, whose termination codon overlaps the start codon of the target gene. In the case of the IL1 alpha gene, the high expression level is largely due to the direct efficient translation initiation at the second cistron, whereas with the IL1ra gene in the same system, the proximal translation initiation region (TIR) provides a high level of coupled expression of the target gene. Thus, pGMCE is a potentially versatile vector for direct prokaryotic expression.

Reagent for introducing pyrene residues in oligonucleotides.

A novel pyrenyl-containing phosphoramidite reagent, N-[4-(1-pyrenyl)butyryl]-O1-(4,4'-dimethoxytrityl)-O2- [(diisopropylamino)(2-cyanoethoxy)phosphino]-3-amino-1 ,2-propanediol (5), has been synthesized from 4-(1-pyrenyl)butanoic acid in four steps with the 52% overall yield and used to incorporate pyrene residue(s) into oligonucleotides. Oligonucleotides 6 and 7, bearing one or two pyrenes at the 5'-terminus, have been prepared by means of that reagent, characterized with fluorescence spectra, and successfully used as primers in a polymerase chain reaction.

A modified approach to identification of the sickle cell anemia mutation by means of allele-specific polymerase chain reaction.

The allele-specific PCR approach has been modified by introducing a second mismatch at the 3'-penultimate link of the primer and used to identify the sickle cell anemia mutation (A-->T transversion in the sixth codon of the human beta-globin gene causing Glu-->Val substitution in the protein), thus obviating the problem of an interpretationally ambiguous 3'-terminal mismatch including T residue.

On inherited fertility in biological systems: a model of correlated fluctuations in the stochastic branching process.

A new evolutionary model with hereditary modes considered as correlated fluctuations of fertility has been proposed. It has been demonstrated that the model allows the global statistical properties of the system to be evaluated, e.g. the ensemble average and the probability of extinction. The results obtained show the increase of instability of a population with the enhancement of inheritance efficiency. The existence of at least an exponential stratification in the population has also been shown. Possible applications of the present model are discussed.

Site-specificity of abnormal excision: the mechanism of formation of a specialized transducing bacteriophage lambda plac5.

Molecular mechanism of the specialized transducing bacteriophage lambda plac5 formation has been studied. Phage-bacterial DNA junctions in lambda plac5 DNA are localized and primary structure of regions of the abnormal excisional recombination leading to the phage formation is elucidated; the crossover region proved to be comparable with the central part of attP and attB sites (the core and the adjacent tetranucleotide) in length and degree of homology. Bacterial insert in lambda plac5 DNA is shown to end immediately after Z-Y spacer, the DNA not containing lacY gene segments. The data obtained led to the conclusion of site-specific (homologous) character of abnormal excision upon formation of lambda transducing bacteriophages. Possible mechanisms of the excision are discussed.

The synthesis of a DNA duplex corresponding to the icosanucleotide sequence at the 5' end of messenger RNA from the gene N of bacteriophage lambda.

In connection with work on the nucleotide sequence of the promoter for the gene N of bacteriophage lambda as well as a study of the mechanism of transcription, a 20-unit long DNA duplex corresponding to the known sequence at the 5' end of the above gene transcript has been synthesized. For synthesis, the required duplex was divided into the following deoxyribooligonucleotides: a) the dodecanucleotide, d-A-T-C-A-G-C-A-G-G-A-C-G (II); b) the octanucleotide, d-C-A-C-T-G-A-C-C- (IV); c) the hexanucleotide, d-G-C-T-G-A-rU (I); and d) dodecanucleotide, d-T-C-A-G-T-G-C-G-T-C-C-T (III). All of the four olignucleotides were chemically synthesized and characterized by extensive chromatographic and fingerprinting methods (after labeling the 5' ends with[32P]phosphate group). Longer polynucleotides (an icosa- and an octadecanucleotide) were prepared by polynucleotide ligase-catalyzed joining of segments I and III and by joining segments II and IV. The use of the octadecanucleotide, d-T-C-A-G-T-G-C-G-T-C-C-T-G-C-T-G-A-rU, in work on the sequence analysis of the promoter is described in the accompanying paper. The octadecanucleotide and icosanucleotide were hybridized together to give the double-stranded duplex.