Gunnar Bjursell

Covalently closed circular duplex DNA of Epstein-Barr virus in a human lymphoid cell line.

Abstract A non-integrated form of Epstein-Barr virus DNA was purified from the Burkitt lymphoma-derived human lymphoid cell line Raji by CsCl density gradient centrifugation and neutral glycerol gradient centrifugation. This intracellular form of the virus DNA sediments at a rate typical of a covalently closed circular DNA molecule of the size of the virus genome in both neutral and alkaline solution. Treatment with low doses of X-rays leads to a discontinuous conversion of the molecules to a form with the sedimentation properties of open circular DNA (a circular duplex molecule containing one or more single-strand breaks). The direct observation of large circular DNA molecules by electron microscopy further confirms the covalently closed circular duplex structure of part of the intracellular viral DNA. Such circular molecules were not detected in corresponding DNA fractions from Epstein-Barr virus-negative human lymphoid cell lines. In ethidium bromide/CsCl density gradient centrifugation experiments, the purified non-integrated virus DNA behaves as twisted, covalently closed DNA circles with the same initial superhelix density as polyoma virus DNA. The latter additional purification technique permits the isolation of intracellular Epstein-Barr virus DNA in > 90% pure form from non-producer cells. The molecular weight of the circular virus DNA from Raji cells, determined by contour length measurements, is the same within experimental error as that of the linear DNA from virus particles.

Apolipoprotein B gene variants are involved in the determination of serum cholesterol levels: a study in normo- and hypelipidaemic individuals☆

We have investigated the frequencies of 3 restriction fragment length polymorphisms (RFLPs) of the apolipoprotein B (apo B) gene in normo- and hyperlipidaemic individuals. In individuals with type III hyperlipidaemia, the allele frequency for the RFLP detected with XbaI was significantly different from the allele frequency in normolipidaemic individuals and in those with other types of hyperlipidaemia. No significant difference in allele frequency was found among these groups for the RFLPs detected with MspI or EcoRI. Within a sample of 62 normolipidaemic individuals, homozygotes for the X2 allele (cutting site) of the XbaI RFLP had a significantly higher serum cholesterol level than homozygotes for the XI allele, with individuals of the genotype X1X2 having an intermediate value (X2X2 mean 5.71 mmol/l, X1X1 mean 4.81 mmol/l, X1X2 mean 5.30 mmol/l). There were also significant differences in serum triglyceride levels in individuals with different XbaI genotypes. In these normolipidaemic individuals there was no correlation between the EcoRI and MspI RFLP genotypes and levels of any serum lipid variable. Information from the XbaI and EcoRI RFLPs was used in conjunction to define apo B haplotypes. These haplotypes are a more precise measure of the genotypic variation, and they explain a greater fraction of the serum cholesterol and triglyceride levels than the single-site polymorphisms considered separately. This study suggests that variations in the gene for apo B are associated with the determination of serum cholesterol and triglyceride levels both in patients with type III hyperlipidaemia and in the normal population.

Molecular cloning and sequence analysis of cDNA encoding lipoprotein lipase of guinea pig.

We have isolated and sequenced cDNA clones covering the entire coding sequence and short flanking regions of guinea pig lipoprotein lipase. The expression cDNA library used was constructed in lambda gt11 with mRNA derived from adipocytes. The deduced amino acid (aa) sequence starts with a stretch of 17 aa interpreted as a leader peptide. The open reading frame continues with 448 aa residues and ends with a TGA stop codon. Combined with previous data this information allows the assignment of domains in the lipase molecule. A likely candidate for the heparin-binding site is a 9-aa stretch containing five positive charges, analogous to the consensus sequence for receptor-binding sites on apolipoproteins E and B. A previously noted homology to pancreatic lipase is extended. Analysis of polyadenylated RNA from several tissues indicated a high level of expression in adipocytes, heart muscle and mammary gland. No lipoprotein lipase mRNA could be detected in liver. Northern blots revealed three major mRNAs with sizes corresponding to 3.8 kb, 3.3 kb and 2.1 kb, respectively. In adipocytes and heart muscle a fourth mRNA, with an estimated size of 4.5 kb, was also detected. Analysis of genomic DNA by Southern blotting indicated a single gene locus coding for lipoprotein lipase. Hence, modification of the primary transcript seems to be involved in the production of the various mRNAs.

Characterization of the human lipoprotein lipase (LPL) promoter: evidence of two cis-regulatory regions, LP-alpha and LP-beta, of importance for the differentiation-linked induction of the LPL gene during adipogenesis.

When preadipocytes differentiate into adipocytes, several differentiation-linked genes are activated. Lipoprotein lipase (LPL) is one of the first genes induced during this process. To investigate early events in adipocyte development, we have focused on the transcriptional activation of the LPL gene. For this purpose, we have cloned and fused different parts of intragenic and flanking sequences with a chloramphenicol acetyltransferase reporter gene. Transient transfection experiments and DNase I hypersensitivity assays indicate that several positive as well as negative elements contribute to transcriptional regulation of the LPL gene. When reporter gene constructs were stably introduced into preadipocytes, we were able to monitor and compare the activation patterns of different promoter deletion mutants at selected time points representing the process of adipocyte development. We could delimit two cis-regulatory elements important for gradual activation of the LPL gene during adipocyte development in vitro. These elements, LP-alpha (-702 to -666) and LP-beta (-468 to -430), contain a striking similarity to a consensus sequence known to bind the transcription factors HNF-3 and fork head. Results of gel mobility shift assays and DNase I and exonuclease III in vitro protection assays indicate that factors with DNA-binding properties similar to those of the HNF-3/fork head family of transcription factors are present in adipocytes and interact with LP-alpha and LP-beta. We also demonstrate that LP-alpha and LP-beta were both capable of conferring a differentiation-linked expression pattern to a heterolog promoter, thus mimicking the expression of the endogenous LPL gene during adipocyte differentiation. These findings indicate that interactions with LP-alpha and LP-beta could be a part of a differentiation switch governing induction of the LPL gene during adipocyte differentiation.

Tissue-specific regulation of guinea pig lipoprotein lipase; effects of nutritional state and of tumor necrosis factor on mRNA levels in adipose tissue, heart and liver

Levels of mRNA for lipoprotein lipase (LPL) in guinea pig epididymal adipose tissue, heart and liver were determined by dot blot analysis of total RNA using a cDNA probe complementary to the coding region, and compared to the LPL activity. For adipose tissue we also measured the incorporation of radioactivity into immunoprecipitable LPL after pulse-labeling with [35S]methionine. LPL activity was 93%, LPL mRNA 82% and LPL synthesis 85% lower in epididymal fat pads from animals fasted for 48 h compared to rigorously fed animals. In contrast, neither LPL activity nor LPL mRNA levels differed in heart. A single dose of tumor necrosis factor (TNF) decreased LPL activity and LPL mRNA in fat pads with no effects in heart. In the liver, TNF caused a marked increase in LPL mRNA levels, which are normally very low. Northern-blot analysis confirmed a previous observation that the patterns of mRNA species differ between heart, in which a 3.8-kb mRNA dominates, and adipose tissue, in which the LPL mRNAs of 3.3 and 2.1 kb occur in similar abundance as the 3.8-kb species.

Nuclear Jak2 and Transcription Factor NF1-C2: a Novel Mechanism of Prolactin Signaling in Mammary Epithelial Cells

ABSTRACT The classical mechanism by which prolactin transduces its signal in mammary epithelial cells is by activation of cytosolic signal transducer and activator of transcription 5 (Stat5) via a plasma membrane-associated prolactin receptor-Janus kinase 2 (Jak2) complex. Here we describe an alternative pathway through which prolactin via Jak2 localized in the nucleus activates the transcription factor nuclear factor 1-C2 (NF1-C2). Previous reports have demonstrated a nuclear localization of Jak2, but the physiologic importance of nuclear Jak2 has not been clear. We demonstrate that nuclear Jak2 regulates the amount of active NF1-C2 through tyrosine phosphorylation and proteasomal degradation. Our data also demonstrate a link between prolactin and p53 as well as the milk gene carboxyl ester lipase through nuclear Jak2 and NF1-C2. Hence, we describe a novel pathway through which nuclear Jak2 is subject to regulation by prolactin in mammary epithelial cells.

Lack of correlation between the apolipoprotein B Xba I polymorphism and blood lipid levels in a Swedish population

The possible connections between the apolipoprotein B (apo B) XbaI polymorphism and the serum levels of total cholesterol, triglycerides, LDL, HDL and apo B have been investigated among 187 randomly selected subjects from Gothenburg, Sweden. The interferences of age and sex on the serum lipoproteins and apo B concentrations were considered. Using multiple regression analysis, we compared the different lipid levels and the levels of apo B with the genotypes X1X1, X1X2 and X2X2 (X1 = without the XbaI restriction site, X2 = with the site), with age and with sex and with those factors combined with each other. A significantly higher concentration of serum cholesterol and LDL among men than among women was found and total serum cholesterol, LDL and apo B were positively correlated with age. The allele frequency of the XbaI polymorphism in the sample was 0.45 for the allele without the XbaI restriction site. No correlation was found between the apo B genotypes and the levels of serum lipoproteins or apo B.

Long regions of single-stranded DNA in human cells.

WHEN DNA is isolated from actively growing animal cells by a variety of standard procedures, 1–2% of the total DNA is recovered in single-stranded form1–5. Pulse labelling experiments have established that the large majority of the single-stranded DNA is not newly synthesised3–5. However, a minor part of this DNA fraction consists of short fragments of newly replicated DNA released by branch migration during isolation6–7. Studies with synchronised human cells have shown that single-stranded DNA is mainly found during the period of DNA synthesis3,4. In agreement with this, a specific antiserum against denatured DNA used for immunofluorescence has revealed the presence of single-stranded DNA in nuclei of a human lymphoid cell line during the S phase but not during the G1 phase8. DNA hybridisation measurements indicate that the single-stranded sequences belong to the non-repetitious fraction of the DNA and that at least part of this material contains sequences transcribed in vivo2. We have developed gentle isolation methods to purify large, covalently closed circular DNA molecules from human lymphocytes transformed by the Epstein–Barr virus (EBV)9. These methods do not cause the conversion of linear double-stranded DNA to single-stranded forms5,9. We have repeatedly observed by electron microscopy long single-stranded stretches of host DNA that initially co-purified with the guanine–cytosine-rich viral episomal DNA. This single-stranded DNA accounted for about 1% of the total cellular DNA in several different lymphoid cell lines derived from Burkitt lymphoma patients or from healthy donors, while the EBV DNA sequences in the same lines only constituted 0.02–0.1% of the total DNA. Clearly, the single-stranded sequences were mainly or exclusively of host origin. Long regions of single-stranded DNA, indistinguishable from those present in EBV-transformed cells, were also found in similar amounts in two EBV-negative human lymphoid cell lines of different origin, BJAB10 and Molt-4 (ref. 11). This report looks at these single-stranded DNA molecules, which are considerably larger than any previously isolated from animal cells.

The isolation of genomic recombinants for the human apolipoprotein B gene and the mapping of three common DNA polymorphisms of the gene--a useful marker for human chromosome 2.

SummaryWe have used four independently isolated cDNA probes for human apolipoprotein B (apo B), to isolate overlapping genomic recombinants for the 3′ portion of the apo B gene. The cDNA clones and a unique fragment from the genomic recombinant have been used to identify the human apo B gene in DNA from a series of roden x human somatic cell hybrids. Our results provide evidence for the assignment of this gene to the short arm of human chromosome 2 (p23-pter). We have used the cDNA probes to identify three common DNA polymorphisms. The first, detected with the restriction enzyme XbaI and our probe pAB4, has a rare allele frequency of 0.48. The other two polymorphisms are detected with the probe pAB3. The enzyme MspI detects at least three alleles, with frequencies of 0.67, 0.16 and 0.15, while that detected with the enzyme EcoRI has a rare allele frequency of 0.12. The relative position of these polymorphisms has been mapped using the genomic recombinants.Investigation of a small number of haplotypes indicares that there is linkage equilibrium between the polymorphisms, which have a total polymorphism information content (PIC) value of more than 0.8. These polymorphisms will provide useful markers for genetic studies on chromosome 2 and for the analysis of the involvement of variants of the apo B gene in the development of hyperlipidaemia.

Nuclear Janus-Activated Kinase 2/Nuclear Factor 1-C2 Suppresses Tumorigenesis and Epithelial-to-Mesenchymal Transition by Repressing Forkhead Box F1

Progression to metastasis is the proximal cause of most cancer-related mortality. Yet much remains to be understood about what determines the spread of tumor cells. This paper describes a novel pathway in breast cancer that regulates epithelial-to-mesenchymal transition (EMT), motility, and invasiveness. We identify two transcription factors, nuclear factor 1-C2 (NF1-C2) and Forkhead box F1 (FoxF1), downstream of prolactin/nuclear Janus-activated kinase 2, with opposite effects on these processes. We show that NF1-C2 is lost during mammary tumor progression and is almost invariably absent from lymph node metastases. NF1-C2 levels in primary tumors correlate with better patient survival. Manipulation of NF1-C2 levels by expression of a stabilized version or using small interfering RNA showed that NF1-C2 counteracts EMT, motility, invasiveness, and tumor growth. FoxF1 was found to be a direct repressed target of NF1-C2. We provide the first evidence for a role of FoxF1 in cancer and in the regulation of EMT in cells of epithelial origin. Overexpression of FoxF1 was associated with a mesenchymal phenotype, increased invasiveness in vitro, and enhanced growth of breast carcinoma xenografts in nude mice. The relevance of these findings is strengthened by the correlation between FoxF1 expression and a mesenchymal phenoype in breast cancer cell isolates, consistent with the interpretation that FoxF1 promotes invasion and metastasis.