Fluorescence in situ hybridization data on distances between defined genomic sequences are used to construct a quantitative model for the overall geometric structure of a human chromosome. We suggest that the large-scale geometry during the Go/G1 part of the cell cycle may consist of flexible chromatin loops, averaging -3 million bp, with a random-walk backbone. A fully explicit, threeparametric polymer model of this random-walk/giant-loop structure can account well for the data. More general models consistent with the data are briefly discussed.A human chromosome is a very large molecule. Its DNA strand has in the order of 100 million base pairs (Mbp) arrayed along its contour and has a Mr of -1011 Da. Quantitative information on mammalian chromosome geometry during the interphase part of the cell cycle is very extensive for scales <0.01 Mbp (1, 2) but not for larger scales. At the level of -0.001-0.01 Mbp the DNA is associated with proteins to form a chromatin fiber "30 nm in diameter (1, 2); at scales of -0.1 Mbp the chromatin may form loops (2). Very little is known numerically about the larger-scale geometry, comprising >3 orders of magnitude (0.1-300 Mbp), where the difficulty of following the chromatin fiber as it winds and twists its way within the interphase cell nucleus has crippled quantitative analyses. Yet, large-scale geometric structure of chromosomes influences essential cellular processes such as DNA replication and transcription (2), as well as many specialized functions such as repair or misrepair of ionizing-radiation-produced DNA damage (3, 4).Recently, van den Engh et at (5) used fluorescence in situ hybridization data to quantify some intermediate-scale properties of interphase chromosomes. These investigators measured physical distances between pairs of fluorescently marked specific DNA sequences on human chromosome 4 in fibroblast cells fixed on microscope slides. The observations were made for cells in the Go/G1 phase of the cell cycle, the period between mitosis and the onset of DNA replication. Probe pairs having genomic separations from -0.1 Mbp to "4 Mbp were analyzed. A major conclusion was that, on scales from 0.1 Mbp to 1.5 Mbp, chromatin geometry corresponds to a simple random walk. Observed deviations from random-walk behavior at larger genomic separations could be explained by a polymer model in which the DNA of any one chromosome is confined to a spherical subvolume of the interphase nucleus (6). An alternative suggestion was that the deviations were due to "giant" loops, several Mbp in length (7) (Fig. 1). For values of genomic separation <1.5 Mbp, the points lie approximately on a straight line (Fig. 1A), corresponding to random-walk behavior for chromatin, in agreement with earlier data (5). Genomic sites separated by 10-190 Mbp also show an approximately linear relation but with a much smaller slope (Fig. 1 B). Moreover, for these large genomic separations, the statistical distribution of distances for a given probe pair also corresponds to random-walk behavior (F...

Abstract. We determined the folding of chromosomes in interphase nuclei by measuring the distance between points on the same chromosome. Over 25,000 measurements were made in G0/G1 nuclei between DNA sequences separated by 0.15-190 megabase pairs (Mbp) on three human chromosomes. The DNA sequences were specifically labeled by fluorescence in situ hybridization. The relationship between mean-square interphase distance and genomic separation has two linear phases, with a transition at N2 Mbp. This biphasic relationship indicates the existence of two organizational levels at scales >100 kbp. On one level, chromatin appears to be arranged in large loops several Mbp in size. Within each loop, chromatin is randomly folded. On the second level, specific loop-attachment sites are arranged to form a supple, backbonelike structure, which also shows characteristic random walk behavior. This random walk/giant loop model is the simplest model that fully describes the observed large-scale spatial relationships. Additional evidence for large loops comes from measurements among probes in Xq28, where interphase distance increases and then locally decreases with increasing genomic separation. SIaORXLY after cell division, the mitotic chromosomes decondense and diffuse into the interphase nucleus. While individual chromosomes cannot be discerned, important processes related to chromosome function take place. Regulatory factors interact with chromatin, DNA is made accessible for transcription, RNA is produced and processed, DNA is replicated, and repairs are made of DNA strand breaks. When the chromosomes reappear for the next mitosis, they have been duplicated and prepared for rapid partitioning over the daughter cells. The complexity of these processes raises many questions about the large-scale organization of chromosomes and how this organization relates to cell function (e.g., Blobel, 1985;Manuelidis and Chen, 1990;Cook, 1991;Lawrence and Singer, 1991;De Boni, 1994).Diverse models, ranging from highly random to highly organized, have been proposed for the higher-order organization of interphase chromatin. These models variously involve irregularly folded fibers (DuPraw, 1965), radial loop structures (Manuelidis and Chen, 1990), giant loops (Ostashevsky and Lange, 1994), semirigid orientation ("Rabl" configuration) (Rabl, 1885;Comings, 1968), or random polymers confined by tethering or external forces (Hahnfeldt et al., 1993). Some models assign to chromo-

By using a fluorescence in situ hybridization technique we revealed that for nine different q-arm telomere markers the positioning of chromosomes in human G(1) interphase nuclei was chromosome size-dependent. The q-arm telomeres of large chromosomes are more peripherally located than telomeres on small chromosomes. This highly organized arrangement of chromatin within the human nucleus was discovered by determining the x and y coordinates of the hybridization sites and calculating the root-mean-square radial distance to the nuclear centers in human fibroblasts. We demonstrate here that global organization within the G(1) interphase nucleus is affected by one of the most fundamental physical quantities-chromosome size or mass-and propose two biophysical models, a volume exclusion model and a mitotic preset model, to explain our finding.

CITED2 (CBP/p300-interacting transactivator with ED-rich tail 2) is a member of the Cited family of nuclear regulators, previously known as mrg1 (melanocyte-specific gene-related gene 1). CITED2 is inducible by varying stimuli including lipopolysaccharide, hypoxia, and cytokines such as interleukin 9 and interferon ␥. Using the immortalized human chondrocyte cell line, C-28/I2, we investigated whether CITED2 could be responsive to mechanical stimuli, and if so, whether CITED2 could mediate shear-driven regulation of matrix metalloproteinase (MMP) genes. The C-28/I2 cells were cultured under flow shear at 1-20 dyn/cm 2 , and the role of CIT-ED2 in regulation of MMPs was examined using the plasmids encoding sense and antisense CITED2 DNA sequences. The results showed that flow shear at 5 dyn/ cm 2 increased CITED2 mRNA and protein levels and down-regulated MMP-1 and MMP-13 mRNA and protein levels as well as enzyme activities. Consistent with the coordinated expression patterns of CITED2 and MMPs, overexpression of CITED2 repressed MMP-1 and MMP-13 mRNA levels and activities, whereas antisense CITED2 plasmids prevented the shear-induced downregulation of MMP expression. Interleukin-1␤ induced the formation of p300-Ets-1 complexes without affecting expression of CITED2. Transforming growth factor-␤ as well as flow shear at 5 dyn/cm 2 stimulated not only the expression of CITED2 but also the association of CIT-ED2 with p300 by dissociating Ets-1 from p300. These results indicate that CITED2 plays a major role in shearinduced down-regulation of MMP-1 and MMP-13 via a transforming growth factor-␤-dependent pathway.Physical stimuli at appropriate intensities are essential for growth and maintenance of bone and joint tissues (1-3). In vivo studies demonstrate that mechanical loading facilitates the strengthening of bone, and flow-induced shear in articular cartilage stimulates a repair response (4, 5). Chondrocytes in cartilage experience a variety of stresses, strains, and pressure that result from normal activities of daily living. Determining how shear stress alters chondrocyte metabolism is fundamental to understanding how to limit matrix destruction and stimulate cartilage repair and regeneration (31). Matrix metalloproteinases (MMPs) 1 are a family of collagen-degrading proteinases whose expression and activities are altered by mechanical stimuli in various cell types (6 -8). In inflammatory joint diseases such as rheumatoid arthritis and osteoarthritis, MMPs are considered pivotal proteinases of cartilage degradation. Because IL-1 and tumor necrosis factor-␣ are known to stimulate the expression and activities of MMPs, cytokine antagonists and receptor-blocking antibodies have been studied as potential agents for blocking cartilage destruction in joint diseases (9 -11). Our recent study using a human synovial cell line showed that gentle mechanical shear with intensity at a few dyn/cm 2 had anti-inflammatory effects and reduced the expression and activities of many MMPs including MMP-1 and MMP-13 (7). The molecula...

No abstract