Fractions of Chinese hamster DNA differing in their content of guanine+cytosine and evidence for the presence of single-stranded DNA

The "Genomic Code": DNA Pervasively Moulds Chromatin Structures Leaving no Room for "Junk"

The chromatin of the human genome was analyzed at three DNA size levels. At the first, compartment level, two "gene spaces" were found many years ago: A GC-rich, gene-rich "genome core" and a GC-poor, gene-poor "genome desert", the former corresponding to open chromatin centrally located in the interphase nucleus, the latter to closed chromatin located peripherally. This bimodality was later confirmed and extended by the discoveries (1) of LADs, the Lamina-Associated Domains, and InterLADs; (2) of two "spatial compartments", A and B, identified on the basis of chromatin interactions; and (3) of "forests and prairies" characterized by high and low CpG islands densities. Chromatin compartments were shown to be associated with the compositionally different, flat and single- or multi-peak DNA structures of the two, GC-poor and GC-rich, "super-families" of isochores. At the second, sub-compartment, level, chromatin corresponds to flat isochores and to isochore loops (due to compositional DNA gradients) that are susceptible to extrusion. Finally, at the short-sequence level, two sets of sequences, GC-poor and GC-rich, define two different nucleosome spacings, a short one and a long one. In conclusion, chromatin structures are moulded according to a "genomic code" by DNA sequences that pervade the genome and leave no room for "junk".

Yeast and mammalian replication intermediates migrate similarly in two-dimensional gels

In the budding yeast, Saccharomyces cerevisiae, DNA replication initiates at specific, discrete chromosomal locations. At each initiation site, a single small replication bubble is generated, which subsequently expands at Y-like replication forks. We wanted to know whether other eukaryotic organisms utilize similar initiation mechanisms. For this purpose, replication intermediates (RIs) from three different organisms (Schizosaccharomyces pombe, Chinese hamster and human) were mixed individually with RIs from S. cerevisiae and then subjected to two-dimensional (2D) gel electrophoresis under conditions known to resolve molecules having different structures. All of the RIs detected by the hybridization probes we used for each organism migrated nearly identically to specific RIs of similar size from S. cerevisiae, implying that the detected RIs from all the studied organisms have very similar structures and may therefore employ the same basic initiation mechanism.

The mouse immunoglobulin kappa light-chain genes are located in early- and late-replicating regions of chromosome 6

The murine immunoglobulin kappa (kappa) light-chain multigene family includes the constant region (C kappa), joining-region genes, and approximately 30 kappa-variable (V kappa) region families. The entire region occupies an estimated 1,000 to 3,000 kilobases, and some V kappa families have been linked by recombinant inbred mapping. The C kappa gene and 14 V kappa families replicated differently among cell lines of lymphoid and nonlymphoid origin. In nonlymphoid cells, the C kappa gene replicated earlier than the V kappa families. A transition from replication during the second third of S phase for the C kappa gene to later replication during S for V kappa families was observed. The V kappa family (V kappa 21) that maps closest to the C kappa gene, replicated during the first half of the S phase; most of the other V kappa families replicated during the second half of S, and some replicated during the last quarter of the S phase. In lymphoid cells, the kappa locus replicated earlier in the pre-B than in the B-cell lines. In one pre-B-cell line, 22D6, the kappa genes examined replicated at the beginning of the S phase. In the B-cell lines, the EcoRI segment containing the transcribed gene replicated near the beginning of the S phase. Other V kappa families replicated within the first two-thirds of S phase. Some linked V kappa families replicated at similar times. In the B-cell lines, a transition from replication at the beginning of S for the transcribed C kappa and V kappa genes and surrounding DNA sequences to later replication for the other V kappa families was observed. However, in contrast to the non-lymphoid cell lines, the replication of this locus occurred predominantly during the first half of S. The kappa locus contains both early- and late-replicating genes, and early replication is usually associated with transcriptional activity. The results are discussed with respect to the organization of transcriptionally active chromatin domains.

The mouse immunoglobulin kappa light-chain genes are located in early- and late-replicating regions of chromosome 6

The murine immunoglobulin kappa (kappa) light-chain multigene family includes the constant region (C kappa), joining-region genes, and approximately 30 kappa-variable (V kappa) region families. The entire region occupies an estimated 1,000 to 3,000 kilobases, and some V kappa families have been linked by recombinant inbred mapping. The C kappa gene and 14 V kappa families replicated differently among cell lines of lymphoid and nonlymphoid origin. In nonlymphoid cells, the C kappa gene replicated earlier than the V kappa families. A transition from replication during the second third of S phase for the C kappa gene to later replication during S for V kappa families was observed. The V kappa family (V kappa 21) that maps closest to the C kappa gene, replicated during the first half of the S phase; most of the other V kappa families replicated during the second half of S, and some replicated during the last quarter of the S phase. In lymphoid cells, the kappa locus replicated earlier in the pre-B than in the B-cell lines. In one pre-B-cell line, 22D6, the kappa genes examined replicated at the beginning of the S phase. In the B-cell lines, the EcoRI segment containing the transcribed gene replicated near the beginning of the S phase. Other V kappa families replicated within the first two-thirds of S phase. Some linked V kappa families replicated at similar times. In the B-cell lines, a transition from replication at the beginning of S for the transcribed C kappa and V kappa genes and surrounding DNA sequences to later replication for the other V kappa families was observed. However, in contrast to the non-lymphoid cell lines, the replication of this locus occurred predominantly during the first half of S. The kappa locus contains both early- and late-replicating genes, and early replication is usually associated with transcriptional activity. The results are discussed with respect to the organization of transcriptionally active chromatin domains.

A destabilized DNA conformation associated with tightly bound nuclear proteins in active genes of rat myoblast

Purified nuclei from tissue cultured myoblasts were disrupted and centrifuged to equilibrium in a sarcosyl-caesium chloride gradient. A small portion (1.3% - 1.9%) of the non histone proteins (NHP) were banded with DNA in a high density region of the gradient. The DNA tightly bound to proteins representing about 0.6% of the total nuclear DNA was degraded after treating cell nuclei with S1 nuclease or DNAse I but resisted to mild micrococcal nuclease digestion. A large portion of the DNA sequences complementary to homologous RNA was concentrated in this DNA-proteins fraction. These finding suggest that a subset of NHP strongly associated to the active DNA regions play a role in the destabilisation of the double helical DNA during transcriptional processes.

Repetitive DNA sequences in methotrexate- and methasquin-sensitive and -resistant Chinese hamster cell lines

DNA purified from a Chinese hamster cell line of lung fibroblast origin (DC83F) was analyzed by density gradient centrifugation and by gel electrophoresis after restriction endonuclease digestion in order to fractionate discrete repetitive fractions within the total DNA. No obvious satellite DNAs were resolved using the CsCl or Ag-Cs2SO4 density gradient conditions described herein. However, analysis of the digestion products of a battery of restriction endonucleases indicated that three of these enzymes, EcoR1, HaeIII, and XhoI, yielded discrete fragments which could be visualized with EtBr staining or identified by scintillation counting of [3H]DNA. DNAs from several highly (greater than or equal to hundredfold increased resistance) antifolate-resistant sublines of DC-3F, characterized by a large homogeneously staining region (HSR) in the chromosome complement, were examined with both techniques and compared to the parental, antifolate-sensitive cell line DNA. The density gradient profiles and electrophoretic patterns of restriction endonuclease digests were identical among all the cell lines examined and were indistinguishable from those of the parental DC-3F DNA.

Isolation of single stranded DNA related to the transcriptional activity of animal cells

Single stranded DNA (s.s.DNA) comprising 1-2% of the total nuclear DNA was isolated by an improved method of hydroxyapatite chromatography from native nuclear DNA3 of embryonic chick cells, labeled for several cell generations with 3H-thymidine. Small quantities of 3H-DNA were annealed with a large excess of unlabeled DNA or polysomal RNA from chick embryos. Hybridization kinetics (monitored by the use of SI nuclease digestion, hydroxyapatite chromatography and thermalfusion), indicated that s.s.DNA belongs to the non repetitious fraction of the cell genome. One third represents DNA sequences engaged in the transcription of messenger RNA's.

Distribution of repetitious sequences in chick nuclear DNA

By an improved method of hydroxylapatite chromatography, the reassociated sequences of chick nuclear DNA were isolated, and their base composition analysed. By increasing the amount of reassociation, the G + C content of the renatured sequences decreased progressively to reach a mean value corresponding to that of the total DNA. In order to study the distribution of the families, or group of families having different amount of reassociation, DNA was fractionated by CsC1 density gradient centrifugation. Fractions having different G + C content were obtained, and their reassociation rates analysed. At high C(o)t value of renaturation (C(o)t=50) the amount of reassociated sequences included in the high or in the low buoyant density DNA fractions was approximately the same, but their G + C content was as expected different. At lower C(o)t values of renaturation (between C(o)t of 0.2 and the C(o)t of 10), the results indicated an heterogeneity of the repeated sequences in the A + T rich DNA fractions, as compared to the G + C rich ones.