A biomolecular implementation of logically reversible computation with minimal energy dissipation.

Energy dissipation associated with logic operations imposes a fundamental physical limit on computation and is generated by the entropic cost of information erasure, which is a consequence of irreversible logic elements. We show how to encode information in DNA and use DNA amplification to implement a logically reversible gate that comprises a complete set of operators capable of universal computation. We also propose a method using this design to connect, or 'wire', these gates together in a biochemical fashion to create a logic network, allowing complex parallel computations to be executed. The architecture of the system permits highly parallel operations and has properties that resemble well known genetic regulatory systems.

Malaria and the cell cycle.

The current model of cell cycle control features a succession of active cyclin-CDK (cyclin-dependent kinase) complexes, where accumulation of each successive cyclin leads to activation of its associated kinase. Cell fusion experiments have shown that nuclei sharing common cytoplasm progress through the cell cycle in synchrony. During schizogony of Plasmodium falciparum, nuclear division occurs asynchronously, and thus cannot be regulated by synthesis and accumulation of cyclins in the cytoplasm. We suggest that schizonts must have a ready pool of cyclins for activating all stages of the cycle, and that the cell cycle is regulated independently in each nucleus.

Genetic variation in mouse apolipoprotein A-IV expression is determined pre- and post-transcriptionally.

Among inbred mouse strains there is a striking genetic variation in the levels of apolipoprotein A-IV (apoA-IV) mRNA in the liver, although intestinal mRNA levels vary only twofold in these strains. In the present study we have characterized the apoA-IV expression phenotypes in strains C57BL/6J and 129/J, and investigated the molecular basis for the genetic variation. We report that the two strains differ eight- to tenfold both in the levels of apoA-IV mRNA and in the rate of apoA-IV protein synthesis in liver. Presumably due to the increased synthetic rate, strain 129 exhibits a threefold higher concentration of apoA-IV protein in the circulation. mRNA synthesis and turnover studies indicate that both transcriptional and post-transcriptional events contribute to the genetic variation in steady state apoA-IV mRNA levels. An analysis of the levels of apoA-IV mRNA derived from 129 and C57BL/6 alleles in F1 mice indicates that the genetic control of apoA-IV mRNA levels involves both cis-acting elements linked to the apoA-IV gene, and genetically distinct trans-acting factors.

Genetic variation in mouse apolipoprotein A-IV due to insertion and deletion in a region of tandem repeats.

We have detected three unique apolipoprotein A-IV (apoA-IV) charge isoforms in strains of commensal mice. The cDNA sequences for one representative of each isoform (Mus domestesticus strains C57BL/6J and 129/J and Mus castaneus) revealed a polymorphism within a series of four imperfect repeats encoding the sequence Glu-Gln-Ala/Val-Gln. Insertions or deletions of 12 nucleotides within this repetitive region have given rise to three genotypes characterized by three (129), four (C57BL/6), or five (M. castaneus) copies of the repeat unit. To ascertain the extent of this variation among other species of the Mus genus, we sequenced this region of apoA-IV cDNAs from eight additional M. domesticus inbred strains and from five wild-derived Mus species. All eight additional M. domesticus strains examined had four repeat units, as found in C57BL/6. Among wild-derived mice, however, one species (Mus spretus) had three repeats, two species (Mus cookii and Mus cervicolor) had four repeats, and two species (Mus hortulanus and Mus minutoides) had five repeats. A lack of correlation between the number of repeat units and the phylogeny of Mus species indicates that independent mutations may have occurred throughout the evolution of specific mouse lineages. We suggest that the repetitive nature of the polymorphic sequence may predispose this region to slippage errors during DNA replication, resulting in frequent deletion/insertion mutations.

cDNA cloning of carboxyl ester lipase from human pancreas reveals a unique proline-rich repeat unit.

We report the isolation and nucleotide sequence of the cDNA for carboxyl ester lipase (CEL) from human pancreas. CEL was purified from human pancreas and microsequence analysis was performed on the amino-terminal and internal peptides. Peptide sequence was used to design oligonucleotide probes for screening a human pancreas cDNA library. Partial length cDNAs for CEL were isolated from the library, and the 5' portion of the cDNA was obtained using the anchored polymerase chain reaction. The deduced amino acid sequence indicates that mature CEL contains 722 amino acids and is synthesized with a 20 amino acid leader peptide. The amino acid sequence is rich in proline (12.2%), with 68% of the proline residues occurring within the final 25% of protein length. This is due to the occurrence of a series of proline-rich tandem repeat units near the carboxyl terminus, and accounts for the previously observed species variation in CEL size and amino acid composition. The primary sequence of CEL shows strong similarity to members of the serine esterase family, including the identical G-E-S-A-G motif at the putative active site. A striking homology also occurs between CEL and acetylcholinesterase and cholinesterase, essential enzymes of the nervous system. Proteins with cholesteryl esterase activity have been detected in extra-pancreatic tissues including liver, intestine, kidney, aorta, macrophage, and in the milk of some species (human, gorilla, cat, dog), but not others (rat, cow). To clarify the structural relationships between these various esterases and CEL, we used the CEL cDNA to study expression in pancreas and liver. CEL mRNA was abundant in pancreas of human and rat, with the human CEL mRNA approximately 300 nucleotides larger than that from rat. CEL mRNA was not detected in human adult or fetal liver, nor in rat liver. These results indicate that CEL is not synthesized in significant amounts in liver, and suggest that the cholesterol esterase activity that has been described in liver may be due to a distinct enzyme, or may be derived from pancreas, as has been proposed for the cholesterol esterase activity in intestine.