Presented at: "Biochemistry of the Cell Nucleus: Mechanism and Regulation of Gene Expression", of the Ninth Meeting of the Federation of European Biochemical Societies, August 25-30, 1974, Budapest, Hungary, and Published in Proc. Ninth FEBS Meeting, Volume 33, Pages 419-428, (1974): 

Division of Medical Oncology, Stanford University School of Medicine, Stanford, California 94305, USA. 

References:
Additional References:
Links:
Feedback: 

² This paper is dedicated to the memory of two inspiring teachers and profound friends, Professors Richard J. Winzler and Alfred E. Mirsky, who pioneered in biochemical studies of the control of leukemic cell differentiation: (1958) Proc. Soc. Exp. Biol. Med. 98: 887; and DNA template activity: (1961) Biochim. Biophys. Acta 47: 130. 

INTRODUCTION

Chromatin is the term employed to describe the complexes found between nuclear DNA and associated non-DNA macromolecules in the living cell (Frenster and Herstein, 1973). The variety of these associated macromolecules, and the intensity of their association with nuclear DNA in the living cell, provide the molecular bases for control of the template activity of DNA during the RNA synthesis of gene transcription and the DNA synthesis of gene replication (Frenster, 1965e).

At least two physical states of chromatin are recognized in living cells, the extended form of active chromatin being termed euchromatin, and the condensed form of repressed chromatin being termed heterochromatin (Frenster, et al 1963; Frenster, 1974). Euchromatin and heterochromatin can each be isolated in parallel from animal cells (Frenster, et al, 1963), and can each be analyzed for its macromolecular composition and its DNA template activity after such isolation (Frenster, 1965b). These analyses have revealed that equal amounts and types of repressor histones are present within euchromatin and heterochromatin, but that the histones within euchromatin are partially displaced from their underlying DNA templates (Fig. 1) by a variety of nuclear polyanionic macromolecules, including nuclear RNA, acidic proteins, saline-soluble proteins, hydrophobic proteins, phosphoproteins, and lipoproteins (Frenster, 1965b).

Fig. 1. Polycationic repressor histones (dark blocks) form electrostatic complexes with the phosphate groups on the exterior of the DNA helix within repressed heterochromatin, thus inactivating the DNA template for RNA or DNA synthesis or for reaction with such DNA ligands as acridine orange. Repressor histones are displaced from such DNA templates by tryptic digestion, or by reaction with such synthetic or natural polyanions as polyethylene sulfonate, phosphoproteins, or nuclear RNA, thus freeing the underlying DNA template for RNA and DNA synthesis and for reaction with acridine orange as a probe molecule (Frenster, 1965b). Specific species of nuclear RNA (de-repressor RNA, chromosomal RNA) appear to be capable of selecting specific portions of the genome for such gene de-repression (Frenster, 1965 b, c, e, f; Holmes et al, 1974a,b).

These nuclear polyanionic de-repressors are thought to combine with the polycationic histone repressors (Frenster, et al, 1961), resulting in the displacement of the histones from theunderlying DNA template, with the subsequent extension and activation of such DNA templates (Frenster, 1965a) for RNA and DNA synthesis. This displacement of repressor histones from DNA templates during gene activity offers the possibility of utilizing as probe molecules those ligands that bind to free DNA (Frenster, 1965f) but are prevented in binding to DNA covered by histone repressors (Frenster, 1971). We have developed high-resolution electron microscopic methods which now permit such molecular probes of chromatin within living human cells (Frenster, 1972). Because changes in gene activity are a regular feature of both cell differentiation (Feinendegan, et al 1964) and cell division (Simmons, et al, 1973), we have applied these high-resolution ultrastructural probes to human bone marrow cells undergoing differentiation and division in-vivo.

RESULTS

Bone marrow spicules (Fig. 2) and peripheral blood cells were isolated under sterile conditions from patients as previously described (Frenster, 1971; Frenster, et al, 1958), and were subjected to the analysis of active DNA templates while in the living state (Frenster, 1972; Nakatsu et al, 1974).

Fig. 2. Electron micrograph of bone marrow spicule isolate from a normal human subject and subjected to reaction with acridine orange while the spicule remained in the living state (Frenster, 1971, 1972). Granulocytic and erythrocytic precursor cells are seen distributed throughout the marrow sinusoids, with more mature cells found within the blood capillary. Magnification: 1,000 X.

Fig 3. Electron micrograph of a granulocyte precursor cell (promyelocyte) within a marrow spicule of a normal human subject after the living marrow spicule was reacted with acridine orange (Frenster, 1971, 1972). The electron-dense reaction products are localized exclusively within the extended euchromatin portion of the cell nucleus, and are very numerous compared to those in cells of more advanced stages of cell differentiation (Table 1). Magnification: 7,500 X.

Fig. 4. Electron micrograph of an erythrocyte precursor cell (proerythroblast), prepared as in Fig. 3. The electron-dense reaction products are localized exclusively within the extended euchromatin portion of the cell nucleus, and are very numerous compared to those in cells of more advanced stages of cell differentiation (Table 1). Magnification: 7,500 X.

---

These probe sites were counted within each cell by one observer, and another observer independently classified the stage of cell differentiation or of cell division (Frenster, et al, 1973), using the ultrastructural criteria previously reported for differentiating granulocytes (Anderson, 1966), differentiating erythrocytes (Bessis and Breton-Gorius, 1962), and dividing cells (Robbins and Gonatas, 1964).

A total of 123 normal differentiating granulocyte precursor cells (Fig. 3) and 176 normal differentiating eryhtrocyte precursor cells (Fig. 4) were analysed. These analyses revealed a progressive decrease during the course of normal cell differentiation in the percentage of cells containing either large or small probe sites (Table 1).

In addition, these analyses revealed a progressive decrease during cell differentiation in the number of large probe sites per positive cell (Table 1) as well as a progressive decrease in the fraction of all probe sites represented by large probe sites (Table 1). These observed decreases in DNA template activity during the course of normal bone marrow cell differentiation are statistically highly significant (P<0.01) by Mann's test for trend significance (Hollander and Wolfe, 1973).

Similarly, a total of 50 dividing normal bone marrow cells were analyzed (Figs. 5, 6).

Fig. 5. Electron micrograph of a mitotic granulocyte precursor cell in the metaphase stage of cell division. The nuclear membrane has disappeared, the nucleolus has segregated, and the chromosomes are highly condensed and individually distinct. No reaction product is visible, even though surrounding non-dividing cells are heavily labelled following reaction of this living bone marrow spicule from a normal subject with acridine orange (Frenster, 1971, 1972). Earlier and later stages of cell division display significantly higher degrees of DNA template activity. Magnification: 7,500 X.

Fio. 6. Electron micrograph of a mitotic granulocyte precursor cell in the anaphase stage of cell division. The nuclear membrane is reforming, the nucleolus is dispersing, and the chromosomes remain highly condensed but are less individually distinct as they migrate to the poles of the new daughter cells. No reaction product is visible, even though surrounding non-dividing cells are heavily labelled following reaction of this living bone marrow spicule from a normal human subject with acridine orange (Frenster, 1971, 1972). Earlier and later stages of cell division display significantly higher degrees of DNA template activity (Table 2). Magnification: 7,500 X.

These analyses revealed a progressive decrease in the number of probe sites per cell through the early and late stages of prophase, to an absence of any probe sites in metaphase, anaphase, and early telophase, with a subsequent rapid increase in the number of probe sites per cell in late telophase, returning to a basal level in interphase. These variations in DNA template activity through the course of cell division corelate directly with a similar variation in the rates of RNA synthesis in each of the corresponding phases of cell division (Simmons, et al, 1973), and suggest that DNA template activity is the critical factor in such synthesis.

Single living human bone marrow cells have been analyzed for DNA template activity by a high-resolution electron microscopic molecular probe technique. DNA template activity was found to decrease through the course of cell differentiation and cell division, indicating the importance of gene repression mechanisms in these cellular processes, and during neoplastic transformation by oncogenic viral genomes which integrate with the euchromatin portion of the host genome. 

The rates of cellular RNA and DNA synthesis decline during cell differentiation (Feinendegan et al, 1964) and during the mitotic stages of cell division (Simons et al, 1973). The molecular mechanisms mediating such a decline may include changes in the size of mononucleotide pools, changes in RNA and DNA polymerase activity, or changes in DNA template activity (Keshgegian et al, 1971), but previous studies have not distinguished among these possibilities. We have applied high-resolution electron microscopic molecular probe analysis to differentiating and dividing cells within human bone marrow, and have demonstrated a close correlation between DNA template activity and the previously-reported variations of RNA and DNA synthetic activity (Nakatsu et al, 1974).

These single-cell analyses of living cells within their native habitat in bone marrow spicules correlate closely with independent studies utilizing isolated chromatin which have revealed a similar direct correlation between DNA template activity as measured by quantitating probe sites within isolated chromatin, and the rate of RNA synthesis by such isolated chromatin when supplied with excess amounts of exogenous RNA polymerase and monoribonucleotides (Seligy and Lurquin, 1973).

These quantitative data suggest that a progressive repression of DNA templates is an important mechanism during normal cell differentiation and cell division in human bone marrow, and now permit an analysis of both cell differentiation and cell division as dynamic throughput systems (Frenster, 1965d). The normal controlling macromolecules for such repression of DNA templates involve the repressor histones (Frenster, 1965b), and these are in turn influenced by a variety of nuclear polyanions, including nuclear RNA, acidic proteins, saline-soluble proteins, hydrophobic proteins, phosphoproteins, and lipoproteins, all of which are found in excess in active chromatin (Frenster, 1965b), and all of which can function as de-repressors of previously-repressed DNA templtes (Frenster, 1965b).

Further, such proteins (Frenster et al, 1960) and RNA (Frenster et al, 1963) are known to be synthesized within the same nuclei which are being in-turn influenced by their activity, thus providing a simple feedback system for the positive control of individual gene activity (Frenster and Herstein, 1973). The recent demonstration that DNA sequences coding for such de-repressor RNA species is interspersed as moderately repetitive sequences (Holmes et al, 1974a) at the 5' terminal of unique DNA sequences coding for structural proteins (Holmes et al, 1974b) strongly supports the concept of such selective de-repressor RNA and corresponding reiterative DNA in controlling selective gene transcription (Britten and Davidson, 1969; Frenster and Herstein, 1973).

In addition to the role of endogenous nuclear RNA species in controlling DNA template activity, the current intensive investigations of oncogenic RNA and DNA viruses strongly suggest that exogenous nucleic acids can also selectively de-repress host DNA template activity (Coggin et al, 1970). The normal human genome is known to posess sequences in common with oncogenic viral nucleic acids (Baxt et al, 1973), and these nucleic acids have been shown to integrate into the host genome in the active euchromatin portion of the cell nucleus (Astrin, 1973; Shih et al, 1973; Hampar et al, 1974). Although it has been presumed that such viral integration is covalent and therefore double-stranded because of the resistance of such integration to alkaline sucrose gradients (Sambrook et al, 1968), it is now recognized that non-covalent Watson-Crick base pairing can survive such alkaline sucrose gradients (Simpson et al, 1973), raising the possibility that oncogenic viral integration is single-stranded (Fig. 7),

Fig. 7. Single-stranded integration of oncogenic viral genome into the host genome (Frenster, 1965f).
(+) strand of the viral genome is non-covalently bound via Watson-Crick base pairing with complementary sequences on the (-) strand of the host genome, allowing messenger RNA synthesis on the host (+) strand and anti-messenger RNA synthesis on a restricted portion of the (-) strand of the viral genome as long as the virus is integrated (Astrin, 1973; Shil et al, 1973). Release of the virus from integration allows copying from both strands of the viral genome (ibid). The synthesis of complementary messenger and anti-messenger RNA has been reported during viral infection (Aloni and Locker, 1973). The single-stranded regions of the viral and host genomes are particularly sensitive to 5-BUDr and/or radiation treatment of the transformed cell (Monkehaus, 1973), allowing easy recovery of virus after release from integration following single-strand breakage (Lowy et al, 1971; Teich et al, 1973). 

The single-stranded integration of oncogenic viral genomes is supported by the separations achieved with gradients after extraction of total DNA from EB-virus transformed cells (Nonoyama and Pagano, 1972; Hampar et al, 1973), and is reminiscent of the non-covalent association of exogenous transforming DNA in Drosophila (Fox et al, 1970) and Arabidopsis (Ledoux and Huart, 1974).

These data concerning the integration of oncogenic viral genomes into the active euchromatin portion of the host genome finally suggest that the induction of single-strand breaks in the unpaired regions of the integration may be a method for inducing reversion of previously-transformed cells by removal from the state of integration of the viral genome, and such a reversion has been reported following treatment of MSV-transformed cells with fluorodeoxyuridine (Nomura et al, 1973).

References:

Aloni Y, and Locker H, (1973), Virology 54: 495.

Anderson DR, (1966), J. Ultrastr. Res. Suppl. 1, 9: 5.

Astrin SM, (1973), Proc. Natl. Acad. Sci. U.S.A., 70: 2304.

Baxt W, Yates JW, Wallace HJ, Holland JF, and Spiegelman S, (1073), Proc. Natl. Acad. Sci. U.S.A., 70: 2629.

Bessis MC, and Breton-Gorius J, (1962), Blood 19: 635.

Britten RJ, and Davidson EH, (1969), Science 165: 349.

Coggin JH, Ambrose KR, and Anderson NG, (1970), J. Immunol. 105: 524.

Feinendegan LE, Bond VP, Cronkite EP, and Hughes WL, (1964), Ann. N.Y. Acad. Sci., 113: 727.

Fox AS, Duggleby WF, Gelbart WM, and Yoon SB, (1970), Proc. Natl. Acad. Sci. U.S.A., 67: 1834.

Frenster JH, (1965a), Nature 205: 1341.

Frenster JH, (1965b), Nature 206: 680.

Frenster JH, (1965c), Nature 206: 1269.

Frenster JH, (1965d), Nature 207: 1139.

Frenster JH, (1956e), Nature 208: 894.

Frenster JH, (1965f), Nature 208: 1093.

Frenster JH, (1971), Cancer Res. 31: 1128.

Frenster JH, (1972), Nature New Biol. 236: 175.

Frenster JH, (1974), in "The Cell Nucleus", Vol. 1, Academic Press, New York, pp. 565-580.

Frenster JH, Allfrey VG, and Mirsky AE, (1960), Proc. Natl. Acad. Sci. U.S.A., 46: 432.

Frenster JH, Allfrey VG, and Mirsky AE, (1961), Biochim. Biophys. Acta 47: 130.

Frenster JH, Allfrey VG, and Mirsky AE, (1963), Proc. Natl. Acad. Sci. U.S.A., 50: 1026.

Frenster JH, Best WR, and Winzler RJ, (1958), Proc. Soc. Exp. Biol. Med. 98: 887.

Frenster JH, and Herstein PR, (1973), New Eng. J. Med. 288: 1224.

Frenster JH, Nakatsu SL, and Masek MA, (1973), Adv. Cell Biol. 3: 1 (Academic Press, New York).

Hollander M, and Wolfe DA, (1973), "Non-Parametric Statistical Methods", Wiley, New York.

Holmes DS, Mayfield JE, and Bonner J, (1974), Biochemistry 13: 849.

Holmes DS, and Bonner J, (1974), Proc. Natl. Acad. Sci. U.S.A., 71: 1108.

Hompar B, Derga JG, Martos LM, Tagamets MA, Chang SY, and Chakrabarty M, (1973), Nature New Biol. 244: 214.

Keshgegian AA, Meisner LF, and Frenster JH, (1971), Proc. Fourth Leukocyte Culture Conf.", Appleton-Century-Crofts, New York, pp. 361-366.

Ledoux L, and Huart R, (1974), Nature 249: 17.

Lowy DR, Rowe WP, Teich N, and Hartley JW, (1971), Science 174: 155.

Monkehaus F, (1973), Int. J. Rad. Biol. 24: 517.

Nakatsu SL, Masek MA, Landrum S, and Frenster JH, (1974), Nature 248: 334.

Nomura S, Fischinger PJ, Mattern CFT, Gerwin BI, and Dunn KJ, (1973), Virology 56: 152.

Nonoyama M, and Pagano JS, (1972), Nature New Biol. 238: 169.

Robbins E, and Gonatas NK, (1964), J. Cell Biol. 21: 429.

Sambrook J, Westphal H, Srinivasan PR, and Dulbecco R, (1968), Proc. Natl. Acad. Sci. U.S.A., 60: 1288.

Seligy VL, and Lurquin PF, (1973), Nature New Biol. 243: 20.

Shih TY, Khoury G, and Martin MA, (1973), Proc. Natl. Acad. Sci. U.S.A. 70: 3506.

Simmons T, Heywood P, and Hodge L, (1973), J. Cell Biol. 59: 150.

Simpson JR, Nagle WA, Bick MD, and Belli JA, (1973) Proc. Natl. Acad. Sci. U.S.A., 70: 3660.

Teich N, Lowy DR, Hartley JW, and Rowe WP, (1973), Virology 51: 163.

"Model of Single-Stranded Integration of Oncogenic Viral Genomes".

"Oncogenes as Molecular Targets within Active Chromatin", (Frenster, 1999).

"Electron Micrographs of Human Lymphocytes before and after Activation by PHA", (Busch, 1974). 

"Mated Models of Gene Regulation in Eukaryotes".