Faculty Bibliography

We determined the nucleotide sequence of a mouse and a human cDNA, which we designate STPK13, that encodes an apparent protein kinase related to that encoded by the Drosophila melanogaster polo gene and the Saccharomyces cerevisiae CDC5 gene. The polo and CDC5 gene products are required for normal mitosis. The STPK13 mRNA is regulated during terminal erythrodifferentiation and during the cell cycle. Within the precommitment period of murine erythroleukemia cell terminal differentiation, most of the poly(A) tail is lost from the STPK13 mRNA, but the body of the mRNA remains unchanged in abundance; this poly(A) loss does not occur in mutant erythroleukemia cells that fail to commit to terminal differentiation. During the cell cycle, the abundance of the body of the STPK13 mRNA fluctuates. The mRNA is present in growing but not in nongrowing cells. It reaches a maximum abundance during G2/M phase, is absent or present at only low levels during G1 phase, and begins to reaccumulate at approximately the middle of S phase. The cell cycle-associated accumulation and loss of the STPK13 mRNA could cause a similar fluctuation in abundance of its encoded protein kinase, thereby providing a maximum amount during M phase, when the kinase is thought to function, and little or none at other times of the cell cycle. Posttranscriptional regulation must be responsible for the cell cycle-associated fluctuations because transcription rates are relatively constant during different times of the cell cycle when there are large differences in mRNA abundance

Monkey COS-7 cells were transformed with BLUR8 DNA, a pBR322 plasmid containing a human Alu-family sequence at the BamHI site. Within 24 hr of transformation 2-5% of the BLUR8 molecules recovered resisted cleavage by Dpn I, indicating they had replicated. Electron microscopy revealed appropriately sized circular molecules with replication bubbles whose centers were mapped to the Alu insert. A 16-base-pair deletion within the Alu sequence prevented replication. The results indicate that certain Alu sequences can serve as origins of replication in COS-7 cells

4.5S RNA is a group of RNAs 90 to 94 nucleotides long (length polymorphism due to a varying number of UMP residues at the 3' end) that form hydrogen bonds with poly(A)-terminated RNAs isolated from mouse, hamster, or rat cells (W. R. Jelinek and L. Leinwand, Cell 15:205-214, 1978; F. Harada, N. Kato, and H.-O. Hoshino, Nucleic Acids Res. 7:909-917, 1979). We have cloned a gene that encodes the 4.5S RNA. It is repeated 850 (sigma = 54) times per haploid mouse genome and 690 (sigma = 59) times per haploid rat genome. Most, if not all, of the repeats in both species are arrayed in tandem. The repeat unit is 4,245 base pairs long in mouse DNA (the complete base sequence of one repeat unit is presented) and approximately 5,300 base pairs in rat DNA. This accounts for approximately 3 X 10(6) base pairs of genomic DNA in each species, or 0.1% of the genome. Cultured murine erythroleukemia cells contain 13,000 molecules per cell of the 4.5S RNA, which can be labeled to equilibrium in 90 min by [3H]uridine added to the culture medium. The 4.5S RNA, therefore, has a short half-life. The 4.5S RNA can be cross-linked in vivo by 4'-aminomethyl-4,5',8-trimethylpsoralen to murine erythroleukemia cell poly(A)-terminated cytoplasmic RNA contained in ribonucleoprotein particles

Others have reported that about two-thirds of the polyadenylated RNA of sea urchin or frog eggs contains short interspersed repetitive sequence transcripts, a much larger proportion than that found in mRNA of somatic cells. Thus, it appears that incompletely processed transcripts accumulate in these oocytes. Also, in what may be a related phenomenon, the nuclear concentration of U1 RNA (involved in processing hnRNA) decreases during growth of frog oocytes. To pursue this question in mammals, Northern blots of RNA from mouse oocytes and eggs collected before and after meiotic maturation were probed with genomic clones containing rodent Alu-equivalent sequences. The Alu sequence is the predominant short interspersed repetitive element in the genome and is abundant in hnRNA. When compared on the basis of mRNA content, the oocyte and egg RNA contained less short repetitive sequence transcripts than liver or brain cytoplasmic RNA. Using a U1 RNA-specific probe, the concentration of U1 RNA in mouse oocyte nuclei was found to be quite similar to that in somatic cells, and U1 RNA was stable during meiotic maturation. These results suggest that processing of transcripts in mouse oocytes does not possess the unusual features observed in lower animals

We have isolated eight cDNA clones complementary to the human Kpn I repeat and determined the base sequence of three. We have also determined a portion of the base sequences of three human Kpn I family members. The three cDNA sequences are extensively homologous with the 3' ends of the three genomic Kpn I family members and with a simian Kpn I family member recently described [Thayer, R. E. & Singer, M. F. (1983) Mol. Cell. Biol. 6, 967-973]. The genomic repeats terminate in regions of sequence rich in dAMP residues close to sequences at the 3' ends of the cDNA clones; a precise 3'-terminal nucleotide cannot be distinguished. These structural features are consistent with the dispersal of at least some Kpn I family members by entry into genomic DNA of copies of Kpn I RNA transcripts. Each cDNA contains a long poly(dAMP) homopolymer at its 3' end and either one or two A-A-T-A-A-A polyadenylylation signal sequences upstream from it, suggesting that Kpn I family members may be transcribed by RNA polymerase II

Approximately 1% of heterogeneous nuclear RNA and approximately 0.035% of cytoplasmic RNA from a cultured line of human lymphoblastoid cells is complementary to a long dispersed repetitious sequence that comprises at least 6% of human DNA. The complementary nuclear RNA is both heterogeneously and discretely sized and is present in both poly(A)-terminated and non-poly(A)-terminated molecules. The complementary cytoplasmic RNA is mainly in discretely sized molecules ranging in size from approximately 600 to 8200 bases, some of which are most abundantly represented in poly(A)-terminated molecules, whereas others are most abundantly represented in non-poly(A)-terminated molecules. Few, if any, of the complementary cytoplasmic RNAs can be found associated with polyribosomes. The dispersed repeat sequence exhibits substantial restriction enzyme fragment length polymorphisms in human DNA and is also present in mouse DNA, although some regions of the human repeat appear to be more abundantly represented in mouse DNA than are other regions

A family of related sequences that includes approximately 500,000 members is the most prominent short dispersed repeat family in primate and rodent DNA's. The primate sequence is approximately 300 base pairs in length and is composed of two imperfectly repeated monomer units, whereas the rodent repeat consists of only a single monomer. Properties of this repeat sequence, its flanking sequences in chromosomal DNA, and RNA's transcribed from it suggest that it may be a mobile DNA element inserted at hundreds of thousands of different chromosomal locations