Chromosome-wide distribution of haplotype blocks and the role of recombination hot spots.

Recent studies of human populations suggest that the genome consists of chromosome segments that are ancestrally conserved ('haplotype blocks'; refs. 1-3) and have discrete boundaries defined by recombination hot spots. Using publicly available genetic markers, we have constructed a first-generation haplotype map of chromosome 19. As expected for this marker density, approximately one-third of the chromosome is encompassed within haplotype blocks. Evolutionary modeling of the data indicates that recombination hot spots are not required to explain most of the observed blocks, providing that marker ascertainment and the observed marker spacing are considered. In contrast, several long blocks are inconsistent with our evolutionary models, and different mechanisms could explain their origins.

Intracellular distribution of actin in cells of the organ of Corti: a structural basis for cell shape and motility.

Immunofluorescence staining and phalloidin labeling have provided localization of actin in the sensory and supporting cells of the inner ear at the light microscopic level. However, with electron microscopy, neither actin nor actin filaments have been found in the outer hair cell body. This paper describes various techniques utilized to preserve and identify cytoplasmic actin at the ultrastructural level. Post-embedding staining of Lowicryl K4M sections, pre-embedding staining of permeabilized cells of the organ of Corti, pre-embedding staining of vibratome sections, and pre-embedding staining of permeabilized dissociated cells documented the presence of actin, but each of these techniques was best suited to localize actin in specific parts of the cell. Cytoplasmic actin was labeled when isolated cells were lightly fixed and membranes were permeabilized with detergent--conditions under which the cell ultrastructure was compromised. Under conditions of optimal fixation, cytoplasmic filaments embedded in the dense granular matrix of the hair cell cytoplasm were observed.

Detection of mRNAs encoding distinct isoenzymes of type II calcium/calmodulin-dependent protein kinase using the polymerase chain reaction.

A modification of the polymerase chain reaction (PCR) was used to amplify nucleotide sequences encoding the 50-kDa (alpha) or 58- to 60-kDa (beta',beta) subunits of a brain-specific type II calcium/calmodulin-dependent protein kinase (CaM kinase II). Rat brain RNA from different regions and at different postnatal ages was purified, and reverse transcriptase was used to produce cDNA templates. Oligonucleotide primer pairs flanking a unique sequence in the coding region of the beta',beta subunit-specific cDNA or a unique sequence in the 3' noncoding region of the alpha subunit-specific cDNA were used to amplify sequences encoding portions of these subunits by PCR. Adult rat forebrain contained approximately three times as much alpha subunit mRNA as beta',beta subunit mRNA, whereas adult rat cerebellum contained a molar ratio of 1 alpha: 5 beta',beta. Intermediate levels of alpha and beta',beta subunit mRNAs were observed in adult pons/medulla, and in 4- and 8-day neonatal forebrain. This amplification assay was also used to demonstrate the presence of alpha subunit mRNA in cerebellar granule cells and 4-day neonatal forebrain, which was reported to be undetectable by other methods. Cerebellar granule cells contained less alpha subunit RNA relative to whole cerebellum, suggesting that this cell type expresses an isoform of CaM kinase II containing less alpha subunit protein in the holoenzyme. The observed levels of subunit-specific mRNAs were shown to parallel the levels of expressed protein subunits, suggesting that expression of kinase isoforms is transcriptionally regulated. The data also indicate that the conditions used for amplification of CaM kinase II mRNAs are semiquantitative.