MICROSCOPY RESEARCH AND TECHNIQUE 40:418–431 (1998) Nuclear Morphology During the S Phase GIOVANNI MAZZOTTI,1* PIETRO GOBBI,1 LUCIA MANZOLI,2 AND MIRELLA FALCONI3 1Istituto Anatomia Umana Normale, Università di Bologna, 40126 Bologna, Italy Morfologia Umana Normale, Università di Chieti, 66013 Chieti, Italy 3Istituto Citomorfologia Normale e Patologica, CNR, c/o Ist. Rizzoli Codivilla-Putti, 40136, Bologna, Italy 2Istituto KEY WORDS DNA replication; S phase; nuclear morphology; BrdU; immuno electron microscopy ABSTRACT In order to evaluate at the ultrastructural level the chromatin arrangement during the S phase of the cell cycle, the detection of Bromodeoxyuridine (BrdU) by immunogold has been performed in synchronized 3T3 fibroblasts, regenerating liver, and Friend Leukemia Cells (FLC). After a 5-minute BrdU pulse, this label is detected in 10-nm-wide fibers, organized as lacework and assumed to be replication units. In the early part of the S phase, DNA replication units are localized exclusively in the dispersed chromatin domains far from the nuclear envelope. In the middle S, replication occurs at the border between condensed and dispersed chromatin and, finally, in late S, it mainly occurs in perinuclear heterochromatin regions. After replication, the 10-nm fibers can condense in heterochromatin without translocation. Chromatin is highly dispersed in early S and computer image analysis shows an increase in condensed chromatin areas ranging from 13 to 18% at the end of the S phase with a temporal and morphological pattern of distribution characteristic for each cell type. Scanning transmission electron microscopy demonstrates a regular and repetitive structure of dispersed chromatin, represented by a ring-like arrangement of the 10-nm fibers; assuming the same spatial distribution, gold particles that identify incorporated BrdU confirm this organization. By evaluating the organization and the distribution of DNA replication units during S phase, the results suggest that DNA replication occurs at a nucleosomal-like fiber level and that replicating enzymes machinery moves over a fixed template. Microsc. Res. Tech. 40:418–431, 1998. r 1998 Wiley-Liss, Inc. INTRODUCTION The functional steps which characterize the cell cycle are related to specific morphological modifications involving mainly the cellular dimensions, the protrusions of the cell surface, and the different levels of chromatin condensation. While mitosis identifies characteristic chromatin patterns, the three-dimensional arrangements of DNA filament during G1 and S phase are still unknown, resulting in a different and or conflicting interpretation from several authors in the last 30 years (Huberman, 1987; Kondra and Ray, 1978; Manuelidis and Chen, 1990; Nurse, 1994). In fact, during the interphase period, the nuclear structure and in particular the nuclear ultrastructure are characterized by significant variations in the proportions of condensed and dispersed chromatin. In Hyacintus orientalis, a continuous increase of chromatin condensation from G1 to metaphase has been described by electron microscopy. In the same study it has been demonstrated that this phenomenon results in species and tissues specificity in plants, insects, or mammals (Nagl, 1980). In the macronucleus of Euplotes, Kluss (1962) and Prescott (1962) have demonstrated a low electron opacity in the zone of DNA synthesis, and this concept was confirmed by a study of Hay and Revel (1963) that described a fine filamentous DNA network in the DNA replicating areas. Similar results are also described in regenerating rat hepatocytes where the dispersed chror 1998 WILEY-LISS, INC. matin appears as a 12-nm thin filament (Derenzini, 1979; Derenzini et al., 1981, 1984). High resolution electron microscopy autoradiography demonstrated that DNA synthesis occurs in dispersed chromatin or at the border zone between condensed chromatin and interchromatin regions (Angelier et al., 1976; Edemberg and Huberman, 1975; Fakan and Bernhard, 1973; Fakan and Hancock, 1974; Kuroiwa, 1974; Williams and Ockey, 1970). In 1982 Gratzner developed a monoclonal antibody that detects 5-bromodeoxyuridine, a synthetic analogue of thymidine that is incorporated into newly synthesized DNA. This very simple method has a very high efficiency, low background, is not time-consuming, and can be applied both in flow cytometry and in light as well as electron microscopy, thus allowing a very complete analysis of the cell cycle (Danova et al., 1988; Moran et al., 1985; Risio et al., 1986; Schutte et al., 1987; Sugihara et al., 1986; Thiry, 1988; Thiry and Dombrowicz, 1988). Despite the possibility that BrdU could induce perturbation of growth and differentiation (Brown and Schildkraut, 1979; Meuth and Green, 1974), it has been demonstrated that the results are similar to those Contract grant sponsor: Italian CNR, PF ACRO; Contract grant sponsor: Ministero della Ricerca Scientifica e Technologica (MURST) Regione Emilia Romagna ISS FONDI 1%. *Correspondence to: Giovanni Mazzotti, MD, Ist. Anatomia Umana Normale, v. Irnerio 48, 40126, Bologna, Italia. Received 21 November 1995; accepted in revised form 12 January 1996 NUCLEAR MORPHOLOGY DURING THE S PHASE obtained with tritiated thymidine (Lin and Allison, 1993; Mazzotti et al., 1990). So multiple DNA replicon domains spatially and temporally organized during the S phase of the mammalian cells (Nakamura et al., 1986) have been demonstrated and computer analysis (Van Dierendonck et al., 1989) or the confocal microscopy of immunostained cells (Neri et al., 1992) show evidence within the S phase of different staining patterns. Electron microscopy studies on synchronous and asynchronous cells allowed a better definition of these events both in vitro and in vivo (Mazzotti et al., 1990; Rizzoli et al., 1992; Vitale et al., 1991). After incorporation of H3 thymidine, using biotinylated dUTP into permeabilized cells or antibodies against 5-bromodeoxyuridine to detect in vivo or in vitro DNA replication, it has been demonstrated that DNA synthesis occurs at fixed sites in close association with the nuclear matrix (Carrı̀ et al., 1986; McCready et al., 1980; Nakayasu and Berezney, 1989; Neri et al., 1992; Smith and Berezney, 1982). In the columnar cell on the crypt base of mouse duodenum El-Alfy and Leblond (1987, 1988, 1989) and El-Alfy et al. (1994), comparing Feulgen-stained semithin Epon sections and autoradiography, have demonstrated that during the transition from G1 to S phase, there is an increase of distinguishable dots and filaments. Despite these results many questions are still unresolved. In particular, we need to know if in a cycling cell it is possible to discriminate G1 and S phase exclusively from the morphological point of view. The first requirement to answer this question is the identification of a correct experimental model in which it should be possible to detect the whole cell cycle starting from a given unquestionable point. Nowadays the system employed to evaluate the S phase is via the detection of labelled precursors of DNA synthesis. In our experience, we have utilized 5-bromo-2-deoxyuridine in synchronized cells both in vitro and in vivo. Synchronization has been induced ‘‘in vitro’’ on 3T3 fibroblasts by the starvation method, and ‘‘in vivo’’ utilizing rat regenerating liver after surgical removal of 3 of the organ. All the samples have been evaluated by ⁄4 transmission electron microscopy and monitored with a cytofluorimeter to exactly evaluate the DNA content of the cell population and the percentage of synchronization, in order to obtain a tight correlation between the different moments (early, middle, and late) of the S phase and nuclear morphology. Moreover, by transmission electron microscopy (TEM) and scanning transmission electron microscopy (STEM), we have analyzed the dimension and the organization of chromatin fibers in replicating DNA. MATERIALS AND METHODS Cell Culture Murine Friend erythroleukemia cells (FLC-745), with a cell cycle of 10–12 hours, were grown in RPMI 1640 medium supplemented with 10% foetal calf serum in a 5% CO2 atmosphere. Swiss 3T3 cells were grown in Dulbecco’s modified minimum essential medium (D-MEM), supplemented with 10% newborn calf serum (NCS). Cells were initially seeded onto flask at a density of 3 3 103/cm2 and usually became confluent 419 after 4 days of culture. The cell synchronization in low serum medium according to Nakayasu and Berezney (1989) was performed incubating them for 72 hours in D-MEM containing 0.5% NCS. In this way, the cells were arrested at G0. Successively, the cells were reseeded at a density of 7 3 103/cm2 onto glass slides in D-MEM containing 10% NCS. Regenerating Rat Liver Sprague Dawley male rats of an average body weight of 180–200 g were used. Two animals were used as controls (Vitale et al., 1991). Partial hepatectomy (about 1 ⁄4 of the liver was left) was performed according to Grisham (1962) and Higgins and Anderson (1931) in order to obtain different times of regeneration before the sacrifice of the animal at 18, 22, 26, 30, and 34 hours. Three rats for each experimental point were sacrificed and the regenerating removed livers processed for electron microscopy. Bromodeoxyuridine Incorporation Exponentially growing FLC were exposed, according to Gratzner (1982), to 5-bromo-2-deoxyuridine (BrdU) at a final concentration of 50 µM, for periods of time ranging from 5 minutes to 12 hours at 37°C. Synchronized Swiss 3T3 fibroblasts at 1-hour intervals from the cell release were exposed to 20 µM BrdU for either 5 or 30 minutes. Control cells, which were cultured for 96 hours in low serum medium, were incubated for 5 or 30 minutes in the presence of BrdU. Alternatively, 13 hours after the block release cells were exposed to BrdU for 30 minutes. One hour before sacrifice Sprague Dawley rats were injected intraperitoneally with 50 mg/kg BrdU. Rat liver nuclei were prepared from regenerating liver by the method of Widnell and Tata (1964), before December 31, 1993. Briefly, after sacrifice, livers were quickly excised, minced with a scalpel, and manual homogenized with 20 strokes in a Dounce pestle type A with a tissue/buffer TM ratio of 1:3 (buffer TM 5 0.25 M sucrose; 0.05 M Tris, pH 7.4; 4.5 mM MgCl2). After filtration through four layers of cheesecloth, the homogenate was centrifuged at 770g for 10 minutes to yield a crude nuclear pellet. The pellet was resuspended in 2.2 M sucrose TM buffer and centrifuged at 40,000g for 90 minutes. This purified nuclear pellet was washed two times and stored in the same buffer at 0°C. Electron Microscopy Murine Friend erythroleukemia cells and nuclear matrices were fixed, after centrifugation at 900 rpm for 5 min, with 1% glutaraldehyde in 0.1 M phosphate buffer, pH 7.4, for 1 hour at room temperature, rinsed three times in 0.15 M phosphate buffer, dehydrated in increasing concentrations of ethanol, and embedded in Epon and London Resin White. Moreover FLC was also embedded in Unicryly (Scala et al., 1992). Swiss mouse 3T3 adherent cells, after two washes in PBS, were fixed as described. The cells were then rinsed three times in 0.15 M phosphate buffer, detached by scraping, and centrifuged at 800 rpm for 10 minutes and embedded in Epon. Regenerating rat livers, 2 3 2 3 1 mm pieces, were processed as described before and embedded in Epon resin. 420 G. MAZZOTTI ET AL. Silver-gold ultrathin sections of all specimens were mounted on nickel grids. Semithin sections (0.1 µm) of Unicryl-embedded FLC cells were deresinated with several washes in acetone and critical point dried. Immunocytochemical Labelling A postembedding immunogold technique for all samples was used as reported by De Mey et al. (1981), involving the following steps: etching of the sections with 10% H2O2 for 10 minutes; incubation with normal goat serum for 30 minutes at room temperature, incubation for different periods of time (2–14 hours for FLC cells and nuclear matrices; 14 hours for 3T3 cells and regenerating livers) at 4°C with an anti-BrdU monoclonal antibody (Becton Dickinson, San Jose, CA) 1 µg/ml in 0.05 M Tris HCl, pH 7.6, containing 0.1% bovine serum (BSA) and incubation with goat anti-mouse IgG conjugated with 15-nm colloidal gold particles (Janssen, Beerse, Belgium) diluted 1:10 in 0.02 M Tris HCl, pH 8.2, containing 0.1% BSA for 1 hour at room temperature. The grids were stained with uranyl acetate and lead citrate and observed employing a Philips (Mahwah, NJ) CM 10 and a Zeiss (Thornwood, NY) EM 109 transmission electron microscope (TEM). After the immunocytochemical labelling, the deresinated sections were coated with a thin platinum-gold layer with a Balzers (Balzers, FL) MED 010 evaporating system and observed with a Philips CM 12 scanning transmission electron microscope (STEM). Computer Image Analysis About 100 TEM pictures with good contrast of each cell type were scanned by a Hewlett Packard (Corvallis, OR) Scanjet and analysed by a Microsoft Image-pro Plus 1.3 PC program in order to obtain, after a repetitive binarization of each image, an automatic recognition of the nuclear envelope and the border between nucleus and nucleolus. With this system it was possible to determinate nuclear and nucleolar diameter and, consequently, the nuclear and nucleolar total area. The exclusion of the nucleolar area permits the estimation of the percentage of the condensed chromatin, computable as dark fields on the digitally acquired image, in comparison to the total nuclear area. Controls The controls consisted, for FLC and 3T3 synchronized fibroblasts, of cells not exposed to BrdU or not incubated with the primary antibody or incubated with a non-immune mouse serum, instead of the primary antibody. For the regenerating rat livers, the following controls were performed: samples obtained from non-hepatectomized rats injected with BrdU or from hepatectomized rats not injected with BrdU. Sections of regenerating liver exposed to BrdU (26 hours after the hepatectomy) were incubated without the primary antibody or with a non-specific immune serum followed by the secondary gold-conjugated antibody. RESULTS 3T3 Fibroblasts Immunocytochemical detection of incorporated BrdU can be performed in TEM without previous acid hydrolysis treatment or proteolytic digestion, as required for light microscopy or flow cytometry, since the antigen is directly exposed on the sectioned surface of the embedding media; in these conditions it is possible to preserve nuclear morphology at the ultrastructural level (Mazzotti et al., 1990). To prevent the possibility of modifications of chromatin organization related to the detachment of the cells after trypsinization and to the consequent modifications of flattened to a spherical shape, 3T3 fibroblasts have been fixed directly in situ onto the culture flask and scraped out. Moreover, some of the samples have been grown on a slide and directly embedded and sectioned as monolayers. Flow cytometer analysis demonstrates that when using the serum deprivation method in G0 phase arrested 3T3 fibroblasts, DNA synthesis takes place 9 hours after the release of the block (Rizzoli et al., 1992). At the 9th hour about 10% of the cells are synthesizing DNA, while at the 13th hour the percentage of cells increases to 70% and exceeds 80% from the 15th to the 20th hour, decreasing to about 50% at the 24th hour. Moreover, the evaluation of the DNA content by propidium iodide (PI) counterstain shows that most of the cells have a 2n DNA content from 9 to 13 hours, suggesting that they could be in early S phase (Rizzoli et al., 1992). With the progression of the DNA synthesis, synchronized cells that have incorporated BrdU show a 3n DNA content from the 15th to the 18th hour, and 4n DNA content from the 21st to 24th hour and they have been interpreted as being, respectively, in the mid and late S phase of the cell cycle (Rizzoli et al., 1992). Immunocytochemical detection of incorporated BrdU in newly synthesized DNA is also detectable 9 hours after the block release and for a pulse of 15–30 minutes demonstrates specific labelling patterns for each of the three periods. In early S, colloidal gold particles are localized exclusively in the interchromatin regions of the dispersed chromatin. The heterochromatin localized at the periphery along the nuclear envelope is not labelled (Fig. 1). Mainly at 9–10 hours but also in the other periods, for short times of incubation with the precursor (1–58) the labelling appears organized in isolated clusters (Fig. 2). At higher magnifications, the gold particles are associated to a regular netting structure of 10–15-nm-wide chromatin fibers organized in a ring-like structure of 100–150 nm in diameter (Fig. 3). From 15 to 18 hours in about 80% of the nuclei, the gold particles are localized at the border between the dispersed and the condensed chromatin associated with the nuclear envelope or the nucleolus (Fig. 4). While in early S phase the inner nuclear regions are characterized by dispersed chromatin, the middle-late S phase is characterized by regularly distributed masses of condensed chromatin about 0.3–0.5 µm in diameter. For the times of incubation with BrdU up to 308, gold particles decorate the periphery of this condensed chromatin, while for longer time periods (more than 608) gold labelling is evident on chromatin clumps, demonstrating the composition of newly synthesized DNA. Twenty-one to 22 hours after the block release, flow cytometry analysis demonstrates that about 80% of the cells contains 4n DNA, so that this period can be considered to be the late S phase of the cells. This stage is characterized by the localization of the gold particles exclusively on condensed chromatin along the nuclear envelope or nucleolus. Dispersed NUCLEAR MORPHOLOGY DURING THE S PHASE Fig. 1. 3T3 fibroblasts. At the early S phase BrdU is detectable in interchromatin regions (arrows). No gold markers are evident at the perinuclear heterochromatin. Bar 5 400 nm. Fig. 2. 3T3 fibroblasts after short pulse (10 min) of BrdU. At the early S phase gold particles are detectable as isolated clusters (p). Bar 5 400 nm. 421 Fig. 3. At higher magnification of the same case as Figure 2 gold particles (W) appear localized in a regular net structure of 10–15-nm chromatin fibers organized in a ring-like structure of 100–150 nm in diameter (arrows). Bar 5 50 nm. Fig. 4. Middle S phase in 3T3 cell. BrdU gold markers appear at this stage at the border between dispersed and condensed chromatin at the nuclear envelope and near the nucleolus (arrows). Bar 5 250 nm. Fig. 5. A late step in the S phase of 3T3 fibroblasts. The absence of colloidal gold particles in the dispersed chromatin is evident as well as the arrangement of heterochromatin in regularly distributed clumps (p). Bar 5 250 nm. Fig. 6. Late S phase after short BrdU pulse, high magnification. Gold markers (●) are arranged in cluster near the heterochromatin masses (p) at the nuclear border (arrows). Bar 5 50 nm. NUCLEAR MORPHOLOGY DURING THE S PHASE chromatin and the inner chromatin masses do not show any gold particle decoration at this stage (Fig. 5). After 58 of incubation with BrdU also at this stage gold particles organized in isolated cluster are detectable. Higher magnifications demonstrate that in these regions too the antibodies coupled to the gold particles detect BrdU that has been incorporated into a 10-nm chromatin fiber (Fig. 6). Regenerating Rat Liver Nuclei Liver regeneration that occurs after the removal of 3⁄4 of the organ represents a very interesting example of in vivo non-malignant proliferation. By 16–18 hours after surgical hepatectomy proliferative activity may be detected in the parenchymal cells surrounding the portal areas. Flow cytometry evaluation of DNA synthesis progression can be obtained observing BrdU incorporation and PI presence on isolated nuclei (Vitale et al., 1991). Eighteen hours after hepatectomy, the parenchymal cells assume a steatosis-like state with diffuse lipid droplets and a large amount of dispersed glycogen granules in the cytoplasm. The chromatin mainly appears to have a dispersed organization, while the condensed one, close to the nuclear envelope, is represented by a very thin layer regularly interrupted in correspondence with the nuclear pore complexes. At this early stage, BrdU detection in newly synthesized DNA shows a labelling pattern regularly distributed on the dispersed chromatin that appears to be represented mainly by a delicate 10–15-nm fiber network where groups of not gold labelled RNP particles are constantly present (Fig. 7). The progression throughout the S phase is characterized by the same staining patterns described in 3T3 fibroblasts with a gold labelling at the boundary of condensed and dispersed chromatin from 23 to 26 hours and on the peripheral condensed chromatin from 26 to 30 hours (Fig. 8). Although in vivo results are more difficult to evaluate as to time and amount of uptake of the precursor than the in vitro models, sometimes, also in regenerating liver nuclei both in the early and in the late stage, isolated cluster of gold particles can be detected both in dispersed and condensed chromatin domains. The condensed chromatin masses of about 0.3–0.5 µm described in the fibroblasts are also present but, in this model, they appear in a more precocious stage, at the end of the early S phase of the cell cycle. Also these masses, at a longer (60 min) time of incubation with BrdU appear intensively decorated by gold particles, demonstrating the presence of newly synthesized DNA (Fig. 9). Nevertheless this last aspect is more clearly detectable in the second half of the S phase. Friend Leukemia Cells (FLC) FLC have been analyzed without synchronization. In a cycling cell population after the administration of BrdU for 58, flow cytometry demonstrates that more than 50% of cells are BrdU positive (Mazzotti et al., 1990). In these labelled cells, electron microscopy detection of the precursor demonstrates gold particles localized in the interchromatic domains in 30% of cells, at the border between inter and heterochromatin in 50% of cells, and on the condensed chromatin in 20% of cells. RNP aggregates of about 0.3 µm are present among the 423 fibers while only some randomly distributed heterochromatin masses can be detected in a single section (Figs. 10 and 11). In all the examined stages and in particular for shorter times (i.e., 5–10 minutes) of incubation with BrdU, it is possible to detect isolated clusters gold particles about 150–200 nm wide both in dispersed and condensed chromatin. In these nuclei, the inner chromatin appears decondensed in all the three observed labelling patterns and there is no morphological evidence of any interchromatin fibers condensation (Figs. 12 and 13). By comparing the labelling patterns and chromatin arrangement distribution, we could assume that in FLC no premature chromosome condensation occurs during the S phase. Computer image analysis performed over about 100 different sections for each cell type shows that the degree of chromatin condensation from early to late S phase increases in all the examined models ranging from about 13% in FLC and 3T3 to 18.2% in hepatocytes (Fig. 14). Despite the different progression with the presence of a peak in middle S for regenerating liver, the final amount of heterochromatin is quite similar in the three different cell types. Finally, to overcome the limitation of the thin section for TEM analysis, in order to identify a three-dimensional organization of chromatin fibers, we utilized STEM analysis. To preserve the spatial architecture of the nucleus, the cells have been embedded in removable resin. After deresination, the cellular structure appears to be recognizable, even also at the higher magnifications (Fig. 15). The method does not preserve the organization of condensed chromatin but, in the dispersed one, it is possible to clearly detect a regular three-dimensional network where fibers of different diameter but mainly of 12–15 nm (including the metal coating of about 3 nm) form meshes of 150–200 nm (Fig. 16). Immunostaining with anti-BrdU in STEM analysis demonstrates that gold particles can assume a ring-like distribution (Fig. 17), limiting areas of 150–200 nm corresponding to the dimensions of the meshes detectable by TEM in dispersed chromatin of the same samples. Moreover, this distribution is similar to the cluster organization of gold particles observed in TEM in all three analyzed cell types. DISCUSSION Despite the fact that the interphase represents the larger parts of the cell cycle, very little is known about the organization of chromatin fibers during this period, especially concerning the DNA replication process. In eucaryotic systems different models and hypotheses have been proposed, but all authors agree about the possibility that DNA synthesis involves initiation processes at multiple sites. However, the amount of those sites has been differently estimated: Edemberg and Huberman (1975) and Hand (1978), by autoradiographic data, suggested that there are about 100,000 initiation sites in a mammalian nucleus, spaced by tracts ranging from 50,000 to 300,000 bp. Recently, in rat fibroblasts incubated with BrdU, Nakamura et al. (1986) observed about 150 foci containing about 20 replication units each. Similar results have been obtained with biotinylated dUTP by Nakayasu and Be- 424 G. MAZZOTTI ET AL. Fig. 7. Early S phase morphology in regenerating rat liver. A regularly distributed pattern of gold particles is detectable (arrows) in the dispersed chromatin. RNP particles without gold labelling are present (p). Bar 5 700 nm. Fig. 8. Late S phase nuclear morphology in regenerating liver. The markers appear at this stage on the condensed perinuclear chromatin masses (arrows). Chromatin clumps are also evident in the nuclear area (p). Bar 5 1 µm. Fig. 9. In the same condition as in Figure 8 but after a longer period of incubation with BrdU, chromatin clumps are also highly labelled by gold particles (arrows). Bar 5 500 nm. NUCLEAR MORPHOLOGY DURING THE S PHASE Fig. 10. In FLC cells BrdU appears distributed in the dispersed chromatin (arrows) in a arrangement similar to that of early S phase; not labelled RNP aggregates (p) are detectable. Bar 5 600 nm. Fig. 11. A ‘‘late S phase’’ pattern of distribution (arrows) is evident in this FLC image. RNP (p) masses are also detectable without gold particles. Bar 5 700 nm. 425 Fig. 12. FLC cells. At high magnification, an ‘‘early S’’ distribution of gold marker for BrdU. No chromatin clumps are evident in the interchromatin regions. Bar 5 400 nm. Fig. 13. No interchromatin clumps are evident, in FLC cells, in a ‘‘late S’’ view. Bar 5 400 nm. 426 G. MAZZOTTI ET AL. Fig. 14. Condensed chromatin area/nuclear surface ratio. Results are expressed as mean percent of condensed chromatin nuclear areas. rezney (1989), Kill et al. (1991), and Tomilin et al. (1995). Independently from the number of the activated units, all these models accept the hypothesis that the initiation sites are linked to physical sites, as nuclear skeleton, nuclear cage, or nuclear matrix (Smith and Berezney, 1982; Smith et al., 1984; Vogelstein et al., 1980). Even if other authors suggest that nuclear skeleton is not the native site of DNA replication in eukaryotic cells (Djondjurov et al., 1986; KrzyzowskaGruca et al., 1983), or that its conformation reflects exclusively the original status of condensation of DNA fibers (Martelli et al., 1991), the main questions about DNA synthesis still concern (1) the level of DNA filament organization that allows the DNA duplication and (2) the localization of the first replication sites. In our experiments, the TEM results demonstrate that, after few minutes of incorporation of BrdU, both in dispersed chromatin and at the border of the condensed one, the chromatin structure appears represented by fibers 10–12 nm in diameter that surround areas about 150–200 nm in diameter. Deresinated samples observed by STEM analysis confirm the threedimensional organization of these 12–15-nm fibers and the presence of gold particles in a ring-like regular distribution. These structures appear the smallest detectable with this method and could be referred to single DNA replication units (Rizzoli et al., 1992). Both by STEM and TEM, these fibers appear as nucleosomic fibers about 10 nm in diameter, which correspond to the first order of DNA condensation (Cocco et al., 1986; De Jong et al., 1990; Maraldi et al., 1979). Recently, by high resolution in situ hybridisation on metaphase chromosomes, we have demonstrated that this kind of fiber represents the lowest order of DNA organization that permits transient DNA denaturation (Rizzi et al., 1995; Rizzoli et al., 1994). Moreover Horowitz et al. (1994) and Woodcock and Horowitz (1995), by electron tomography analysis of chromatin fibers, have proposed a new concept of chromatin fibers architecture represented by open structures of a continuously variable zigzag nucleosomal ribbon. This hypothesis is alternative to the supercoiled solenoid model actually strongly conflicting (Cook, 1995) but it offers a more open substrate to the transcriptional regulatory machinery. This more dynamic process is in agreement with the demonstration of the possibility of the transition between the two forms by changing ionic strength in vitro and in vivo also in reversible mode (Giannasca et al., 1993; Labhart et al., 1981; Lepault et al., 1980; Thoma et al., 1979). All these findings also suggest that DNA replication could occur at nucleosomal-like fiber level. The localization of the initiation sites after DNA precursor administration has been described in the interchromatin domains (Belmont et al., 1989; Brown and Schildkraut, 1979; Carrı̀ et al., 1986; Fakan, 1978; Kay et al., 1971; Mazzotti et al., 1990; Rizzoli et al., 1992), at the boundary of condensed and dispersed chromatin (Babu and Verma, 1987), or close to the nuclear envelope with peripheral localization of both initiation and termination sites (Nicolini et al., 1986). However, the importance of nuclear envelope on replication process has been clearly demonstrated (Mills et al., 1989; Newport, 1987; Sheean et al., 1988). Swiss 3T3 fibroblasts synchronized by starvation method and surgically induced liver regeneration represent, respectively, in vitro and in vivo, the models for the study of S phase progression avoiding the use of chemicals or drugs that could interfere with enzymes or molecules involved in DNA replication. In both these cell lines, the incorporation of BrdUrd in early S phase occurs on dispersed chromatin in the inner regions of the nucleus. In middle S phase DNA replication units are activated at the border between condensed and dispersed chromatin tight to the nuclear envelope and to the nucleolus and in late S exclusively on perinuclear condensed chromatin. Unsynchronized cycling HeLa cells show the same distribution of the staining patterns, thus confirming that DNA replication units localization is specific for each step of the S phase, starting from the inner and moving to the peripheral chromatin domains. Moreover, in synchronized 3T3 fibroblasts we have demonstrated that no translocation of the newly synthesized DNA occurs during S phase. In fact, cells that have incorporated BrdU for 308, 13 hours after metabolic block release, and were cultured until the 23rd hour showed a labelling in the interchromatin domains as in cells collected at 13 hours, immediately after the administration of BrdU (Rizzoli et al., 1992), while the cells that in the same conditions received BrdUrd at the 23rd hour showed the labelling exclusively at the periphery of the nucleus. This observation suggest that chromosomes maintain fixed sites in the nucleus during S phase and that they can replicate different tracts at different times. In 1993 Hozak et al. described replication factories fixed in the inner part of the nucleoskeleton and in cells NUCLEAR MORPHOLOGY DURING THE S PHASE Fig. 15. Deresinated FLC at STEM analysis. Nuclear morphology with condensed areas (p) in opposition to decondensed ones (arrows). Bar 5 200 nm. Fig. 16. Higher magnification of the same samples as in Figure 14. Evident is the three-dimensional architecture of the decondensed 427 chromatin, arranged as a network with meshes of 150–200 nm (p). Bar 5 100 nm. Fig. 17. Backscattered STEM image of the samples in Figure 14 allows detection of the gold particle markers distribution in a ring-like array of about 200 nm. Bar 5 45 nm. 428 G. MAZZOTTI ET AL. incubated with biotinylated dUTP, and suggested that the template moves through these structures. In 1994 Hozak et al. demonstrated that some replication occurs outside these factories at discrete sites of the nucleus on the diffuse skeleton; it becomes significant by mid S phase and later becomes concentrated beneath the lamina. These last results are more in agreement with our data according to the hypothesis that the replication machinery moves with respect to the template. The demonstration of several replication sites activated at different times in the nuclear domains is consistent with the possibility that the DNA synthesis is a continuous process during all the S phase. Van Dierendonk et al., in 1989, evaluating BrdU incorporation, divided S phase in five staining patterns, suggesting that spatial and temporal organization of DNA synthesis seems to be characterized by at least three successive stages of replications. BrdU incubated in situ nuclear matrices also revealed five different staining patterns (Neri et al., 1992). By electron microscopy, the same precursor allows the identification of three or five specific aspects (Mazzotti et al., 1990; Rizzoli et al., 1992). El-Alfy and colleagues (1988, 1994), by comparing H3 thymidine autoradiography and Feulgen reaction, identified four stages characterized by dots and filaments having different sizes. Nevertheless, all these aspects were obtained after short pulses with the precursor, generally below 308. For longer times of incubation, we can observe a continuous growing and diffusion of the labelling around the initiation sites that form structures looking like granules clustered into ring-like, horseshoe-like, and rope-shaped arrays (Nakamura et al., 1986). By considering also that transitional patterns have been observed, Neri et al. (1992) suggested the hypothesis of a dynamic nature of DNA synthesis that is continuous through the S phase. The last question concerns the condensation of the newly synthesized DNA to the final metaphasic chromosomal state. In all the cell models analyzed, the inner chromatin is completely decondensed in early S phase. In fact, in 3T3 fibroblasts, regular masses of condensed chromatin appear in middle-late S phase, in rat liver, in early-middle S phase, while in FLC the correlation of the labelling patterns demonstrates that heterochromatin increase mainly in middle S. From middle to late S with short pulses of BrdU we observed the labelling at the border between condensed and dispersed chromatin; after 1 hour administration or more the gold particles also cover the condensed chromatin; this is particularly evident in synchronized fibroblasts in the inner masses of condensed chromatin. This suggests that DNA replication occurs in dispersed chromatin and, later on, the newly synthesized DNA is assembled in a more condensed form. In regenerating liver, different hepatocyte populations have been provided to perform DNA synthesis at different times 18 hours after hepatectomy. In parenchymal cells the inner chromatin masses are regularly covered by gold particles if administration of BrdU for more than 1 hour occurs during middle-late S, while some of them appear without labelling for BrdU pulse during the first half of the S phase. This observation could explain the peak obtained in regenerating rat liver nuclei by image analysis at middle S, suggesting that all heterochromatin domains are not necessarily represented from condensed DNA fibers. In fact, chromatin condensation may be the result of poly (ADPribose) polymerase activity that transfers the adenosine diphosphate ribose moiety of nicotinamide adenin dinucleotide (NAD) to the nuclear proteins and in particular to H1 histone (Saldit-Geogieff et al., 1980). Both the terminal region of the homopolymer can link to chromosomal proteins, inducing a chromatin condensation that cannot be differentiated by electron microscopy from chromosome fiber condensation (Berger et al., 1979; Farina et al., 1979; Nagl, 1980). This enzyme is activated in proliferating tissues (Menegazzi et al., 1988, 1991); in mouse liver poly (ADP) ribose polymerase transcripts can be found 8 hours after partial hepatectomy and increase up to sixfold within 1–2 days. Both the amount of mRNA and enzymatic activity return to their basal values with the decline of the DNA synthesis (Menegazzi et al., 1990). Moreover, molecules like histones are synthesized all along the S phase (Epner et al., 1981; Kondra and Ray, 1978); while Raska et al. (1990) and Hassan et al. (1994) have demonstrated that transcription and replication occur in complementary regions of the nucleus; Hozak et al. (1994) demonstrated that in structures named nuclear factories, transcription and replication occur simultaneously and can be discriminate exclusively by immunolocalization of the precursors. Recently, in mouse duodenal crypt cell, El Alfy et al. (1995) have been able to divide the cell cycle into 11 stages and have demonstrated that nucleofilaments are assembled in compacted structures that increase in size from the beginning to the end of the S phase. The condensation of newly synthesized DNA could prevent the possibility that the same tract of the genome is duplicated more times during S phase (Babu and Verma, 1987; Harland, 1981). Moreover, it has been demonstrated that the DNA accessibility to various fluorochromes, drugs, or DNase treatment changes during the cell cycle and in particular during S phase (Evenson et al., 1986; Fujiwara, 1967; Pederson and Robbins, 1972; Zamai et al., 1993). For all this evidence we can confirm that S phase is characterized by modifications of the chromatin arrangement, which is very dispersed in early S phase and more condensed in the second part of the S phase. At the moment there are still two classical morphological and histochemical techniques: the detection of precursor of DNA synthesis and the evaluation of the DNA content. This information can be obtained simultaneously with Feulgen reaction combined with H3 thymidine autoradiography (El-Alfy et al., 1994) or by considering that Feulgen hydrolysis denaturation of DNA in fixed cells allows measurement of DNA content, tritiated thymidine, and BrdU incorporation in the same cell (Lin and Allison, 1993). One of the next goals of morphological science in this field is represented by the specific localization of genes in the interphasic nucleus for the evaluation of the exact time of expression and duplication, the position within the nucleus, and the evaluation of gene translo- NUCLEAR MORPHOLOGY DURING THE S PHASE cation in the different conditions (Yokota et al., 1995). In fact, it has been demonstrated that the translocation of specific oncogenes can completely change the proliferating cell rate in B cells lymphomas (Dalla-Favera et al., 1983). The localization of molecular probes combined with morphological and histochemical data will provide new insights for basic and diagnostic purposes in order to evaluate, at the single cell level, the growth kinetic of heterogeneous cell populations in solid tumours, or to predict the effects of gene transplantation in non-proliferating cells as in nervous tissues or in precursors of cycling elements, as stem cells of the bone marrow. ACKNOWLEDGMENTS The authors thank Mr. Aurelio Valmori for photographic assistance. This work was supported by grants from the Italian CNR, PF ACRO, and by 40 and 60% grants from the Ministero della Ricerca Scientifica e Tecnologica (MURST), Regione Emilia Romagna, ISS FONDI 1%. REFERENCES Angelier, N., Bouteille, M., Charret, R., Curgy, J.J., Delain, E., Fakan, S., Geuskens, M., Guelin, M., Lacroix, J.C., Laval, M., Steinert, G., and Van Assel, S. (1976) Detection of nucleic acids by means of ultrastructural autoradiography. J. Microsc. Biol. Cell., 27:215–230. Babu, A., and Verma, R.S. (1987) Chromosome structure: Euchromatin and heterochromatin. Int. Rev. Cytol. 108:1–60. Belmont, A.S., Braunfeld, M.B., Sedat, J.W., and Agard, D.A. (1989) Large-scale chromatin structural domains within mitotic and interphase chromosomes in vivo and in vitro. Chromosoma, 98:129–143. Berger, N.A., Petzold, S.J., and Berger, S.J. (1979) Association of poly(ADP-rib) synthesis with cessation of DNA synthesis and DNA fragmentation. Biochim. Biophys. Acta, 564:90–104. Brown, E.H., and Schildkraut, C.L. (1979) Perturbation of growth and differentiation of friend murine erythroleukemia cells by 5-bromodeoxyuridine incorporation in early S phase. J. Cell. Physiol., 99:261–278. Carrı̀, M.T., Micheli, G., Graziano, G., Pace, T., and BuongiornoNardelli, M. (1986) The relationship between chromosomal origins of replication and the nuclear matrix during the cell cycle. Exp. Cell Res., 164:426–436. Cocco, L., Martelli, A.M., Billi, A.M., Matteucci, A., Vitale, M., Neri, L.M., and Manzoli, F.A. (1986) Changes in nucleosome structure and histone H3 accessibility. Exp. Cell Res., 166:465–474. Cook, P.R. (1995) A chromomeric model for nuclear and chromosome structure. J. Cell Sci., 108:2727–2935. Dalla-Favera, R., Martinotti, S., Gallo, R., Erikson, J., and Croce, C. (1983) Translocation and rearrangements of the c-myc oncogene locus in human undifferentiated B-cell lymphomas. Science, 219:963–997. Danova, M., Riccardi, A., Brugnatelli, S., Fiocca, R., Girino, M., Villani, L., Giordano, M., Buttini, R., Ucci, G., and Mazzini, G. (1988) In vivo bromodeoxyuridine incorporation in human gastric cancer: A study on formalin-fixed and paraffin-embedded sections. Histochem. J., 20:125–130. De Jong, L., Van Driel, R., Stuurman, N., Meijne, A.M.L., and Van Renswoude, J. (1990) Principles of nuclear organization. Cell Biol. Int. Rep., 14:1051–1071. De Mey, J., Moeremans, H., Geuens, G., Nujdsen, R., and De Brabander, M. (1981) High resolution light and electron microscope localization of tubulin with the IGS (immunogold staining) method. Cell. Biol. Int. Rep., 5:889–899. Derenzini, M. (1979) Fine structure of chromatin as visualized in thin sections with the Gautier selective stain for DNA. J. Ultrastruct. Res., 69:239–248. Derenzini, M., Pession-Brizzi, A., and Novello, F. (1981) Relationship between ribonucleoprotein particles containing heterogeneous RNA and ultrastructure and function of chromatin in purified rat hepatocyte nuclei. J. Ultrastruct. Res., 77:66–82. Derenzini, M., Viron, A., and Puvion, E. (1984) Structural organization of rat hepatocyte chromatin as visualized in thin frozen sections selectively stained for DNA. Eur. J. Cell Biol., 33:148–156. 429 Djondjurov, L., Ivanova, E., Markov, D., Bardarov, S., and Sachsenmaier, W. (1986) Is the nuclear matrix the site of DNA replication in eukaryotic cells? Exp. Cell Res., 164:79–96. Edemberg, H.J., and Huberman, J.A. (1975) Eukaryotic chromosome replication. A. Rev. Genet., 9:254–284. El-Alfy, M., and Leblond, C.P. (1987) Long duration of mitosis and consequences for the cell cycle concept, as seen in the isthmal cells of the mouse pyloric antrum. I. Identification of early and late steps of mitosis. Cell Tissue Kinet., 20:205–213. El-Alfy, M., and Leblond, C.P. (1988) Visualization of chromosome assembly during the S and G2 stages of the cycle and chromosome disassembly during the G1 stage in semithin sections of mouse duodenal crypt cells. Am. J. Anat., 183:45–56. El-Alfy, M., and Leblond, C.P. (1989) An electron microscopic study of mitosis in mouse duodenal crypt cells confirms that the prophasic condensation of chromatin begins during the DNA-synthesizing (S) stage of the cycle. Am. J. Anat., 186:69–84. El-Alfy, M., Turner, J.P., Nadler, N.J., Liu, D.F., and Leblond, C.P. (1994) Subdivision of the mitotic cycle into eleven stages, on the basis of the chromosomal changes observed in mouse duodenal crypt cell stained by the DNA-specific feulgen reaction. Anat. Rec., 238:289–296. El-Alfy, M., Liu, D.F., and Leblond, C.P. (1995) DNA changes involved in the formation of metaphase chromosomes, as observed in mouse duodenal crypt cells stained by osmium-ammine. Anat. Rec., 242: 433–448. Epner, E., Rifkind, R.A., and Marks, P.A. (1981) Replication of a and b globin DNA sequences occurs during early S phase in murine erythroleukemia cells. Proc. Natl. Acad. Sci. U.S.A., 78:3058–3069. Evenson, D., Darzynkiewicz, Z., Jost, L., Janca, F., and Ballachey, B. (1986) Changes in accessibility of DNA to various fluorochromes during spermatogenesis. Cytometry, 7:45–53. Fakan, S. (1978) High resolution autoradiography studies on chromatin functions. In: The Cell Nucleus, Vol. V. Academic Press, New York, pp. 3–53. Fakan, S., and Bernhard, W. (1973) Nuclear labelling after prolonged 3H-uridine incorporation as visualized by high resolution autoradiography. Exp. Cell Res., 79:431–444. Fakan, S., and Hancock, R. (1974) Localization of newly-synthesized DNA in a mammalian cell as visualized by high resolution autoradiography. Exp. Cell Res., 83:95–102. Farina, B., Menella, M.R.F., and Leone, E. (1979) Nucleic acids, histones and spermiogenesis: The poly (adenosine diphosphate ribose) polymerase system. In: Macromolecules in the Functioning Cell. Salvatore, Narino, Volpe Eds., Plenum Press, New York, pp. 283–300. Fujiwara, Y. (1967) Role of RNA synthesis in DNA replication of synchronized populations of cultured mammalian cells. J. Cell. Physiol., 70:291–300. Giannasca, P.J., Horowitz, R.A., and Woodcock, C.L. (1993) Transitions between in situ and isolated chromatin. J. Cell Sci., 105:551– 561. Gratzner, H.G. (1982) Monoclonal antibody to 5-Bromo- and 5-Iododeoxyuridine: A new reagent for detection of DNA replication. Science, 218:474–475. Grisham, J.W. (1962) A morphologic study of deoxyribonucleic acid synthesis and cell proliferation in regenerating rat liver: Autoradiography with thymidine-H. Cancer Res., 22:842–849. Hand, R. (1978) Eucaryotic DNA: Organization of the genome for replication. Cell, 15:317–325. Harland, R. (1981) Initiation of DNA replication in eukaryotic chromosomes. Trends Biochem. Sci., 6:71–74. Hassan, A.B., Errington, R.J., White, N.S., Jackson, D.A., and Cook, P.R. (1994) Replication and transcription sites are colocalized in human cells. J. Cell Sci., 107:425–434. Hay, E.D., and Revel, J.P. (1963) The fine structure of the DNP component of the nucleus. An electron microscopic study utilizing autoradiography to localize DNA synthesis. J. Cell Biol., 16:29–51. Higgins, G.M., and Anderson, R.M. (1931) Experimental pathology of the liver. I. Restoration of the liver of the white rat following partial surgical removal. Arch. Pathol., 12:186–202. Horowitz, R.A., Agard, D.A., Sedat, J.W., and Woodcock, C.L. (1994) The three-dimensional architecture of chromatin in situ: Electron tomography reveals fibers composed of a continuously variable zig-zag nucleosomal ribbon. J. Cell Biol., 125:1–10. Hozak, P., Hassan, A.B., Jackson, D.A., and Cook, P.R. (1993) Visualization of replication factories attached to a nucleoskeleton. Cell, 73:361–373. 430 G. MAZZOTTI ET AL. Hozak, P., Jackson, D.A., and Cook, P.R. (1994) Replication factories and nuclear bodies: The ultrastructural characterization of replication sites during cell cycle. J. Cell Sci., 107:2191–2202. Huberman, J.A. (1987) Eukaryotic DNA replication: A complex picture partially clarified. Cell, 48:7–8. Kay, R.R., Haines, M.E., and Johnston, I.R. (1971) Late replication of the DNA associated with the nuclear membrane. FEBS Lett., 16:233–236. Kill, I.R., Bridger, J.M., Campbell, K.H.S., Maldonado-Codina, G., and Hutchinson, C.J. (1991) The timing of the formation and usage of replicase clusters in S-phase nuclei of human diploid fibroblasts. J. Cell. Sci., 100:869–876. Kluss, B.C. (1962) Electron microscopy of the macronucleus of Euplotes eurystomus. J. Cell Biol., 13:462–465. Kondra, P.M., and Ray, M. (1978) Analysis of DNA replication patterns of human fibroblast chromosomes. The replicon map. Hum. Genet., 43:139–147. Krzyzowska-Gruca, S., Zborek, A., and Gruca, S. (1983) Distribution of interchromatin granules in nuclear matrices obtained from nuclei exhibiting different degree of chromatin condensation. Cell Tissue Res., 231:427–437. Kuroiwa, T. (1974) Fine structures of interphase nuclei. Exp. Cell Res., 83:387–398. Labhart, P., Thoma, F., and Koller, Th. (1981) Structural changes of soluble rat liver chromatin induced by shift in pH from 7 to 9. Eur. J. Cell Biol., 25:19–27. Lepault, J., Bram, S., Escaig, J., and Wray, W. (1980) Chromatin freeze fracture electron microscopy: A comparative study of core particles, chromatin, metaphase chromosomes, and nuclei. Nucleic Acids Res., 8:265–278. Lin, P., and Allison, D.C. (1993) Measurement of DNA content and of tritiated thymidine and bromodeoxyuridine incorporation by same cells. J. Histochem. Cytochem., 41:1435–1439. Manuelidis, L., and Chen, T. (1990) A unified model of eukaryotic chromosomes. Cytometry, 11:8–25. Maraldi, N.M., Capitani, S., Cocco, L., and Manzoli, F.A. (1979) Histones assembly and their structural role for nucleosome core. In: Chromatin Structure and Function, part B. C. Nicolini, ed. Plenum Press, New York, pp. 371–387. Martelli, A.M., Falcieri, E., Gobbi, P., Manzoli, L., Gilmour, R.S., and Cocco, L. (1991) Heat-induced stabilization of the nuclear matrix: A morphological and biochemical analysis in murine erythroleukemia cells. Exp. Cell Res., 196:216–225. Mazzotti, G., Rizzoli, R., Galanzi, A., Papa, S., Vitale, M., Falconi, M., Neri, L.M., Zini, N., and Maraldi, N.M. (1990) High-resolution detection of newly synthesized DNA by anti-bromodeoxyuridine antibodies identifies specific chromatin domains. J. Histochem. Cytochem., 38:13–22. McCready, S.J., Godwin, J., Mason, D.W., Brazell, I.A., and Cook, P.R. (1980) DNA is replicated at the nuclear cage. J. Cell Sci., 46:365– 386. Menegazzi, M., Carcereri De Prati, A., Ledda-Columbano, G.M., Columbano, A., Uchida, K., Miwa, M., and Suzuki, H. (1990) Regulation of poly (ADP-ribose) polymerase mRNA levels during compensatory and mitogen-induced growth of rat liver. Arch. Biochem. Biophys., 279:232–236. Menegazzi, M., Gelosa, F., Tommasi, M., Uchida, K., Miwa, M., Sugimura, T., and Suzuki, H. (1988) Induction of poly (ADP-ribose) polymerase gene expression in lectin-stimulated human T lymphocytes is dependent on protein synthesis. B.B.R.C., 156:995–999. Menegazzi, M., Grassi-Zucconi, G., Carcereri De Prati, A., Ogura, T., Poltronieri, P., Nyunoya, H., Shiratori-Nyunoya, Y., Miwa, M., and Suzuki, H. (1991) Differential expression of poly (ADP-ribose) polymerase and DNA polymerase b in rat tissues. Exp. Cell Res., 197:66–74. Meuth, M., and Green, H. (1974) Induction of a deoxycytidineless state in cultured mammalian cells by bromodeoxyuridine. Cell, 2:109–112. Mills, A.D., Blow, J.J., White, J.G., Amos, W.B., Wilcock, D., and Laskey, R.A. (1989) Replication occurs at discrete foci spaced throughout nuclei replicating in vitro. J. Cell Sci., 94:471–477. Moran, R., Darzynkiewicz, Z., Staiano-Coico, L., and Melamed, M.R. (1985) Detection of 5-Bromodeoxyuridine (BrdUrd) incorporation by monoclonal antibodies: Role of the DNA denaturation step. J. Histochem. Cytochem., 35:821–827. Nagl, W. (1980) Condensed chromatin: Species-specificity, tissuespecificity, and cell cycle-specificity, as monitored by scanning cytometry. In: Proceedings of NATO Advanced Study Institute on Cell Growth, Erice, Sicily, pp. 171–221. Nakamura, H., Morita, T., and Sato, C. (1986) Structural organizations of replicons domains during DNA synthetic phase in the mammalian nucleus. Exp. Cell Res., 165:291–297. Nakayasu, H., and Berezney, R. (1989) Mapping replication sites in the eucaryotic cell nucleus. J. Cell Biol., 108:1–11. Neri, L.M., Mazzotti, G., Capitani, S., Maraldi, N.M., Cinti, C., Baldini, N., Rana, R., and Martelli, A.M. (1992) Nuclear matrixbound replication sites detected in situ by 5-bromodeoxyuridine. Histochemistry, 98:19–32. Newport, J. (1987) Nuclear reconstruction in vitro: Stages of assembly around protein-free DNA. Cell, 48:205–217. Nicolini, C., Belmont, A.S., and Martelli, A. (1986) Critical nuclear DNA size and distribution associated with S phase initiation. Peripheral location of initiation and termination sites. Cell Biophys., 8:103–117. Nurse, P. (1994) Ordering S phase and M phase in the cell cycle. Cell, 79:547–550. Pederson, T., and Robbins, E. (1972) Actinomycin binding in synchronized HeLa cells. J. Cell Biol., 55:322–327. Prescott, D.M. (1962) Symposium: Synthetic processes in the cell nucleus. II. Nucleic acid and protein metabolism in the macronuclei of two ciliated protozoa. J. Histochem. Cytochem., 10:145–153. Raska, I., Ochs, R., and Salamin-Michel, L. (1990) Immunocytochemistry of cell nucleus. Electron Microsc. Rev., 3:301–353. Risio, M., Coverlizza, S., Poccardi, G., Candelaresi, G.L., and Gaiola, O. (1986) In vitro immunohistochemical localization of S-phase cells by a monoclonal antibody to bromodeoxyuridine. Basic Appl. Histochem., 30:469–477. Rizzi, E., Falconi, M., Rizzoli, R., Baratta, B., Manzoli, L., Galanzi, A., Lattanzi, G., and Mazzotti, G. (1995) High-resolution FEISEM detection of DNA centromeric probes in HeLa metaphase chromosomes. J. Histochem. Cytochem., 43:413–419. Rizzoli, R., Baratta, B., Maraldi, N.M., Falconi, M., Galanzi, A., Papa, S., Vitale, M., Rizzi, E., Manzoli, L., and Mazzotti, G. (1992) DNA synthesis progression in 3T3 synchronized fibroblasts: A high resolution approach. Histochemistry, 97:181–187. Rizzoli, R., Rizzi, E., Falconi, M., Galanzi, A., Baratta, B., Lattanzi, G., Vitale, M., Manzoli, L., and Mazzotti, G. (1994) High resolution detection of uncoated metaphase chromosomes by means of field emission scanning electron microscopy. Chromosoma, 103:393– 400. Salditt-Geogieff, M., Harpold, M., Sawicki, S., Nevins, J., and Darnell, J.E. Jr. (1980) Addition of poly(A) to nuclear RNA occurs soon after RNA synthesis. J. Cell Biol., 86:844–848. Scala, C., Cenacchi, G., Ferrari, C., Pasquinelli, G., Preda, P., and Manara, G.C. (1992) A new acrylic resin formulation: A useful tool for histological, ultrastructural, and immunocytochemical investigations. J. Histochem. Cytochem., 40:1799–1804. Schutte, B., Reynders, M.M.J., Van Assche, C.L.M.V.J., Hupperets, P.S.G.J., Bosman, F.T., and Blijham, G.H. (1987) An improved method for the immunocytochemical detection of bromodeoxyuridine labelled nuclei using flow cytometry. Cytometry, 8:372–376. Sheean, M.A., Mills, A.D., Sleeman, A.M., Laskey, R.A., and Blow, J.J. (1988) Steps in the assembly of replication-competent nuclei in a cell-free system from Xenopus eggs. J. Cell Biol., 106:1–12. Smith, H.C., and Berezney, R. (1982) Nuclear matrix-bound deoxyribonucleic acid synthesis: An in vitro system. Biochemistry, 21:6751– 6761. Smith, H.C., Puvion, E., Buchholtz, L.A., and Berezney, R. (1984) Spatial distribution of DNA loop attachment and replication sites in the nuclear matrix. J. Cell Biol., 99:1794–1802. Sugihara, H., Hattori, T., and Fukuda, M. (1986) Immunohistochemical detection of bromodeoxyuridine in formalin-fixed tissues. Histochemistry, 85:193–195. Thiry, M. (1988) Immunodetection microscope localization of bromodeoxyuridine incorporated into DNA of ehrlich tumor cell nucleoli. Exp. Cell Res., 179:204–213. Thiry, M., and Dombrowicz, D. (1988) Anti-bromodeoxyuridine monoclonal antibody: An alternative tool for the identification of replicated DNA at the electron microscope level. Biol. Cell., 62:99–102. Thoma, F., Koller, T.H., and Klug, A. (1979) Involvement of histone H1 in the organization of the nucleosome and of the salt-dependent superstructures of chromatin. J. Cell Biol., 83:403–427. Tomilin, N., Solovjeva, L., Krutilina, R., Chamberland, C., Hancock, R., and Vig, B. (1995) Visualization of elementary DNA replication units in human nuclei corresponding in size to DNA loop domains. Chromosome Res., 3:32–40. Van Dierendonck, J.H., Keyzer, R., Van de Velde, C.J.H., and Cornel- NUCLEAR MORPHOLOGY DURING THE S PHASE isse, C.J. (1989) Subdivision of S-Phase by analysis of nuclear 5-bromodeoxyuridine staining patterns. Cytometry, 10:143–150. Vitale, M., Rizzoli, R., Mazzotti, G., Zamai, L., Galanzi, A., Rizzi, E., Manzoli, L., Matteucci, A., and Papa, S. (1991) Characterization and cell cycle kinetics of hepatocytes during rat liver regeneration: In vivo BrdUrd incorporation analysed by flow cytometry and electron microscopy. Cell Prolif., 24:331–338. Vogelstein, B., Pardoll, D.M., and Coffey, D.S. (1980) Supercoiled loops and eucaryotic DNA replication. Cell, 22:79–85. Widnell, C.C., and Tata, J.R. (1964) A procedure for the isolation of enzymatically active rat liver nuclei. Biochem. J., 92:313–321. Williams, C.A., and Ockey, C.H. (1970) Distribution of DNA replicator 431 sites in mammalian nuclei after different methods of cell synchronization. Exp. Cell Res., 63:365–372. Woodcock, C.L., and Horowitz, R.A. (1995) Chromatin organization re-viewed. Trends Cell Biol., 5:272–277. Yokota, H., van den Engh, G., Hearst, J.E., Sachs, R.K., and Trask, B.J. (1995) Evidence for the organization of chromatin in megabase pair-sized loops arranged along a random walk path in the human G0/G1 interphase nucleus. J. Cell Biol., 130:1239–1249. Zamai, L., Papa, S., Lattanzi, G., Rizzoli, R., Rizzi, E., Mazzotti, G., and Vitale, M. (1993) Chromosomal DNA after fixation is less susceptible to denaturation during metaphase than interphase. Eur. J. Histochem., 37:49–58.