John H. Frenster and Paul R. Herstein
Division of Oncology
Stanford University School of Medicine
Stanford, California 94305 

Normal embryogenesis, organ regeneration, neoplastic transformation, immune lymphocyte activation, and the cellular response to steroid hormones all involve activation of DNA template sites for RNA synthesis. Such gene de-repression is mediated by a variety of polyanionic nuclear ligands to DNA, the most significant of which appears to be species of de-repressor RNA. The removal of de-repressor RNA from specific gene sites offers a mechanism for selective feedback inhibition after excessive transcription at such gene sites.

1. Introduction:

One of the central problems of cell physiology is the mechanism whereby the individual animal develops and maintains the diversity of cell types characteristic of the adult state out of the original single-cell fertilized ovum. Although particular adult cell types are often stable for the lifetime of the individual, such phenotypic diversity and stability is not based on a corresponding genetic diversity among these cell types. Rather, cellular and molecular techniques of ever-increasing sensitivity and resolution have revealed that most if not all diploid cells within a normal adult have an identical complement of DNA [1], while at the same time these cells display a diverse expression of such DNA in tissue-specific patterns of RNA synthesis [1, 2]. Such variable gene expression among cells possessing identical gene contents indicates the importance of gene control mechanisms, which permit significant tissue specialization and yet maximize tissue renewal and coordination for the benefit of the parent organism.

Theoretically, variable gene expression could be achieved either by selective gene repression within an otherwise fully expressed genome, or conversely, by selective gene de-repression within an otherwise fully repressed genome. Detailed ultrastructural, biochemical, and metabolic studies have revealed that selective gene de-repression is physiologically the more significant gene control mechanism in animals [3]. As a consequence, particular attention is being paid to the study of those circumstances in animal life during which de-repression of previously-repressed genes has been shown to occur (Table 1).

 

TABLE 1. Natural occurrence of gene de-repression:

(1) Activation of sperm genome in early embryo

(2) Organ regeneration and hypertrophy

(3) Chemical and viral oncogenesis

(4) Steroid hormone activation of target cells

(5) Activation of immune lymphocytes

2. Occurrence of gene de-repression:

During the course of normal embryonic development, a progressive restriction is noted in the diversity of RNA molecules being synthesized, ranging from an extensive transcription of the oocyte genome during the lampbrush stage of oogenesis in the unfertilized oocyte [4] to the very marked restriction of RNA synthesis in the mature nucleate erythrocyte [2] or to the absence of any RNA synthesis at all in the mature sperm [4]. However, when such a non-transcribing mature sperm hybridizes with the mature ovum during fertilization, a very rapid de-repression of the sperm genome occurs, in what is probably the most crucial example of gene de-repression in animal systems [4]. Such de-repression of the paternal genome converts the haploid ovum into a functional diploid cell, and permits an equal paternal contribution to gene expression in all resulting cells of the embryo and the adult [4].

During the course of embryogenesis in the developing liver, a progressive restriction in the diversity of RNA types is observed in the adult liver as compared to the embryonic liver [5], but when the adult liver is induced to regenerate following partial hepatectomy, de-repression of previously-repressed genes again allows reappearance of those RNA species characteristic of the embryonic state [5]. A similar de-repression of previously-repressed genes also is noted if neoplastic cells are assayed for the diversity of RNA species as the neoplasm progresses from a benign nodule to a spontaneous neoplasm to a transplantable neoplasm [6]. Such de-repression of normally-repressed RNA species has also been demonstrated within human leukemic lymphocytes [7]. A similar de-repression of fetal genes may account for the re-appearance within adult human human neoplasms (Table 2) of cell surface antigens characteristic of normal fetal life [8],

 

TABLE 2. Human neoplasms displaying specific fetal antigens:

Adenocarcinoma of colon

Primary hepatoma of liver

Hodgkin's lymphoma

Chronic lymphocytic leukemia

and quite similar de-repression of fetal antigens has been observed in both chemical-induced [9] and viral-induced [10] animal neoplasms. In this regard, de-repression of normally-repressed viral oncogenes is the central feature of the oncogene hypothesis of neoplastic transformation [11].

Detailed studies of the mechanism of action of steroid hormones indicates that each steroid hormone, including estrogen, testosterone, progesterone, aldosterone, and cortisol, can be shown to bind to a corresponding specific receptor protein found in the cytoplasm of cells responsive to that steroid [12]. Following such binding, the hormone-receptor complex is transported into the nucleus, interacting there with the chromatin complex [13], and increasing the synthesis of tissue-specific messenger RNA species [13].

Finally, within immune lymphocytes undergoing activation by specific antigens or by defined mitogens, marked stimulation of the synthesis of heterogenous nuclear RNA is noted [14]. If these stimulated RNA species are found to be absent from non-stimulated cells, then the process of lymphocyte activation can be viewed as involving gene de-repression, and that is already suggested by the character of ultrastructural changes observed during lymphocyte activation [15].

3. Mechanism of gene de-repression:

Because gene de-repression occurs so widely under natural conditions (Table 1), a variety of physiological systems are now available for defining the molecular mechanisms of gene de-repression. These studies have taken advantage of the fact that repressed DNA sequences, found in condensed heterochromatin [16], can be separated from de-repressed DNA sequences, found in extended euchromatin [16], and these can be separately analyzed within their native chromatin complexes by standard biochemical, biophysical, and metabolic methods [3, 17]. Such separated chromatin complexes can also be used as assay systems for those molecular species suspected of possessing repressor or de-repressor activity [ 3]. These studies have uniformly revealed that polycationic histone proteins serve as non-specific repressors of template activity within associated DNA molecules (Fig. 1),


Fig. 1: Interaction of histone repressors with DNA templates during gene repression and de-repression. Under conditions of gene repression, polycationic histone repressors (dark bars) form strong electrostatic bonds with phosphate groups on the exterior of the closed DNA helix [3]. Experimentally, hydrolysis of such histone repressors by proteolytic enzymes such as trypsin, or physical displacement of such histone repressors by strong synthetic polyanions such as polyethylene sulfonate, results in de-repression of the underlying DNA template and increased rates of RNA synthesis [3]. Under physiologic conditions, nuclear polyanions such as phosphoproteins, acidic proteins, and nuclear RNA are capable of similarly displacing histone repressors and increasing the rates of RNA synthesis [3]. Certain species of nuclear RNA appear capable of recognizing specific base sequences on the DNA template during such gene de-repression [3], and thus are strong candidates as the molecular species which effects locus-specific and strand-specific gene de-repression in animals [19]. 

Certain species of nuclear RNA appear to possess the ability to select particular portions of the genome for specific-gene locus de-repression [3], perhaps by virtue of their ability to hybridize [18] to complementary base sequences on the anti-coding strand of the DNA template [19], thereby freeing the coding strand of DNA for messenger RNA synthesis that is both locus-specific and strand-specific [19]. During such gene de-repression, only a small length of DNA, 4-5 base pairs in length, is in the strand-separated state [20]. Recent evidence indicates that de-repressor RNA species can be isolated and purified for further characterization and analysis in assay systems [21], and may form double-stranded RNA-RNA duplexes with the RNA of the immediate transcription product [22].

Because the displacement of histones from underlying DNA templates during gene de-repression results in a conformational change within the chromatin complex [18], it has become possible to develop probes of such complexes with defined molecular species which bind to DNA only if histones are first displaced [23]. Acridine orange is a useful probe in this regard (Fig. 2) because it binds specifically to that DNA which is free of repressor histones [23], and because it can be visualized at high-resolution by electron microscopy to reveal the exact points of gene de-repression within the nuclei of intact human leukemic bone marrow cells [23] and of living human lymphocytes [24].


Fig. 2: Acridine orange probe of sites of gene de-repression within a myelocyte from the bone marrow of an untreated patient with chronic myelocytic leukemia [23]. The electron-dense reaction product localizes over DNA template sites that are free of histone repressors [23]. These sites are found to be confined to the active extended euchromatin portion of the cell nucleus [23] (light-staining nuclear areas in the above electron micrograph). No reaction product is observed in the cytoplasm, the nucleolus, or in the repressed condensed heterochromatin portion of the cell nucleus. Quantitative counts of individual reaction product grains per cell permit an estimation of the degree of gene de-repression for each cell within a cell population. X 9,000. 

The recognition that steroid hormones penetrate into the cell nucleus where they may effect gene de-repression in sensitive tissues [13] indicates to some degree the complexity of molecular interactions in the process of gene de-repression. This analysis can be extended by comparing those nuclear ligands which increase RNA synthesis with those which inhibit such synthesis (Table 3).

 

TABLE 3. Correlation of nuclear ligand binding and RNA synthesis:

Nuclear ligand	Preferred form of DNA	Effect on RNA synthesis
Histones	Double-stranded	Decreased
Protamines	Double-stranded	Decreased
Actinomycin D	Double-stranded	Decreased
Acridine orange	Double-stranded	Decreased
Chloroquine	Double-stranded	Decreased
Lac repressor	Double-stranded	Decreased
Testosterone	Single-stranded	Increased
Estradiol	Single-stranded	Increased
Methylcholanthrene	Single-stranded	Increased
RNA polymerase	Single-stranded	Increased
De-repressor RNA	Single-stranded	Increased
Polyoma viral DNA	Single-stranded	Increased

Such diverse nuclear ligands as estrogens, teststerone, methylcholanthrene, RNA polymerase, de-repressor RNA, and polyoma DNA all increase RNA synthesis, and all bind preferentially to single-stranded DNA [25]. Conversely, such diverse nuclear ligands as histones, protamines, actinomycin D, acridine orange, chloroquine, and lac repressor all decrease RNA synthesis, and all bind preferentially to double-stranded DNA [25]. These correlations strongly support the model of gene de-repression [19] which predicts that nuclear ligands preferring single-stranded DNA will increase RNA synthesis by stabilizing DNA in the open loop conformation favorable to gene transcription [25], while nuclear ligands preferring double-stranded DNA will decrease RNA synthesis by stabilizing DNA in the closed helix conformation unfavorable to gene transcription [25].

4. Control of gene de-repression:

The constraints posed by both the gene-locus specificity and the long-term stability of states of gene de-repression in adult animals provide insight into the nature of the physiological systems which control gene de-repression. Somatic cell genetic analysis has revealed that all but one of the X chromosomes in adult mammalian cells are repressed [26], that the decision which X chromosome will remain active is made early in embryonic life and is then maintained with high fidelity through more than 50 generations of daughter cells [27], and that the control mechanisms are exquisitely sensitive not only to the number of X chromosomes in the cell, but perhaps also to the gene content of each X chromosome [26]. These analyses strongly suggest that feedback mechanisms reflecting messenger RNA accumulation are impinging upon and probably controlling RNA production during selective gene de-repression [26].

Recent studies on the molecular features of messenger RNA synthesis strongly indicate that a significant portion of the immediate transcription product is cleaved (Fig. 3) in forming the final messenger RNA molecule [28], and that the non-messenger fragment may correspond to operator RNA with base sequences complementary to de-repressor RNA of the same gene locus [29].


Fig. 3: Postulated feedback control of gene transcription in animals. Recent studies of the molecular mechanisms of messenger RNA synthesis in animals indicate that a significant portion of the immediate transcription product is cleaved in forming the final messenger RNA (mRNA) molecule [28]. The non-messenger portion of the transcription product includes operator (oRNA), transcribed from the operator locus (o) on the DNA template [29] just prior to transcription of the structural gene (sg) which codes for messenger RNA. De-repressor RNA (dRNA) is thought to bind to the anti-coding strand of the DNA templateduring gene de-repression [19], and therefore to be complementary in base sequence to operator RNA [29]. Such base sequence complementarity would allow RNA-RNA duplex formation between the oRNA and dRNA of a specific gene locus during conditions of oRNA excess, that is, after excessive gene transcription, and would diminish such gene transcription by removing dRNA from the operator site on the DNA template [29]. RNA-RNA duplexes could include the messenger RNA portion (heterometric duplex RNA) or could exclude the mesenger RNA portion (homometric duplex RNA). Both types of RNA-RNA duplexes have recently been isolated from a variety of normal animal tissues [22]. 

In this regard, the ability of oncogenic viruses to de-repress fetal genes in the animal cell genome [10] may reflect either the ability of viral RNA to act as a fraudulent type of de-repressor RNA (Fig. 3), or the ability of viral DNA to insert in or near the operator locus of the host genome (Fig. 3) , a condition postulated in the protovirus hypothesis of neoplastic transformation [30].

5. Clinical implications:

A review of the circumstances in which gene de-repression normally occurs (Table 1) gives some suggestion of the possible clinical significance of the emerging data concerning gene de-repression in man [31].

An understanding of gene de-repression in the early developing embryo may well give us insight into the relationships of viral infections and chemical teratogens in producing anomalies in the course of normal organ development, while an understanding of the control of gene de-repression during organ regeneration and hypertrophy opens the possibility of organ replacement via regeneration in the adult patient.

The increasing data linking de-repression of fetal genes (Table 2) with the natural history of human neoplasms [8], as well as the well-established data in which de-repression of fetal genes can be documented during chemical [9] and viral [10] oncogenesis in animals, offers both insight into the neoplastic process as well as clinical opportunities to approach the therapy [31], diagnosis [8], and prevention [8] of human neoplasms by immunologic means. The recent recognition that immuno-therapeutic human lymphocytes [31, 32] may be activated via gene de-repression mechanisms [31] while the neoplastic target cells may themselves have already undergone a separate gene de-repression [31] offers some indication of the scope and complexity of biological interactions soon to be encountered at the clinical level, as does the intrinsic molecular complexity of the human genome as revealed by recent studies [33-35]. 

Supported in part by Grants CA-10174 and AM-01006 from the National Institutes of Health, and by a Research Scholar Award from the Leukemia Society to Dr. Frenster. 

1. McCarthy BJ, and Hoyer BH, Proc. Natl. Acad. Sci. U.S.A., 52: 915 (1964).

2. Grouse L, Chilton MD and McCarthy BJ, Biochemistry 11: 797 (1972).

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

4. Davidson EH, "Gene Activity in Early Development", (Academic Press, New York, 1968).

5. Church RB, and McCarthy BJ, J. Mol. Biol. 23: 477 (1967).

6. Turkington RW, Cancer Res. 31: 427 (1971).

7. Neiman PE, and Henry PH, Biochemistry 8: 275 (1969).

8. Gold P, Ann. Rev. Med. 22: 85 (1971).

9. Stonehill EH, and Bendich A, Nature 228: 370 (1970).

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

11. Todaro GJ, and Huebner RJ, Proc. Natl. Acad. Sci. U.S.A., 69: 1009 (1972).

12. Spelsberg TC, Steggles AW, and O'Malley BW, J. Biol. Chem. 246: 4188 (1971).

13. Liao S, Liang T, and Tymoczko JL, Nature New Biol. 241: 211 (1973).

14. Cooper HL, J. Biol. Chem. 243: 34 (1968).

15. Tokuyasu K, Madden SC, and Zeldis LJ, J. Cell Biol. 39: 630 (1968).

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

17. McConaughy BL, and McCarthy BJ, Biochemistry 11: 998 (1972).

18. Dahmus ME, and Bonner J, Fed. Proc. 29: 1255 (1970).

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

20. Saucier JM, and Wang JC, Nature New Biol. 239: 167 (1972).

21. Holmes DS, Mayfield JE, Sander G, and Bonner J, Science 177: 72 (1972).

22. Kronenberg LH, and Humphreys T, Biochemistry 11: 2020 (1972).

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

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

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

26. Grumbach MM, Morishima A, and Taylor JH, Proc. Natl. Acad. Sci. U.S.A., 49: 581 (1963).

27. Davidson RG, Nitowsky HM, and Childs B, Proc. Natl. Acad. Sci. U.S.A., 50: 481 (1963).

28. Melli M, and Pemberton RC, Nature New Biol. 236: 172 (1972).

29. Herstein PR, and Frenster JH, In: "Proc. Sec. Conf. on Embryonic and Fetal Antigens in Cancer", (National Technical Information Service, U.S. Dept. of Commerce, 1972, Springfield, VA), pp. 3-5.

30. Temin HM, Proc. Natl. Acad. Sci. U.S.A., 69: 1016 (1972).

31. Frenster JH, and Rogoway WM, In: "Fifth Annual Leukocyte Culture Conf.", Ed. Harris JE, (New York, Academic Press, 1970), pp. 359-371.

32. Cheema AR, and Hersh EM, Cancer 29: 982 (1972).

33. Saunders GF, Shirakawa S, Saunders PP, Arrighi FE, and Hsu TC, J. Mol. Biol. 63: 323 (1972).

34. Frenster JH, In: "The Cell Nucleus", Ed. Busch H, vol. 1, (Academic Press, New York, 1974), pp. 565-580.

35. Frenster JH, Nakatsu SL, and Masek MA, In: "Advances in Cell and Molecular Biology", Ed. DuPraw EJ, vol. 3, (Academic Press, New York, 1974), pp. 1-19.

1. DeCarvalho S, "Effect of RNA from Normal Human Marrow on Leukaemic Marrow In-Vivo", Nature 197: 1077-1080 (March 16, 1963).

2. Frenster JH, and Herstein PR, "Gene De-Repression", New Eng. J. Med. 288: 1224-1229 (June 7, 1973).

3. Niu MC, and Segal SJ, eds., "The Role of RNA in Reproduction and Development", North-Holland Publishing Co., Amsterdam-London, (1973). pp. 1-357.

4. Herstein PR, and Frenster JH, "Mated Models of Gene Regulation in Eukaryotes", in: "Embryonic and Fetal Antigens in Cancer", vol. 2, pp. 5-7, (Anderson NG, and Coggin JH, eds.), National Technical Information Service, U.S. Dept. Commerce, Springfield, Virginia, 1972.

5. Frenster JH, "Oncogenes as Molecular Targets within Active Chromatin", AACR, Nov. 1999.

6. Frenster JH, "Yeast  RNA  Re-Programming  of  Already-Active  Mammalian Chromatin", RNA2002, p. 592, (June, 2002).

7. Gottesfeld JM,  and Barbas CF III, "RNA as a Transcriptional Activator".

8. Frenster JH, and Hovsepian JA, "Overshoot in Late Telophase for RNA Re-Programming of Mitotic Chromatin".

9. Hovsepian JA, and Frenster JH, "RNA-Induced Melting of DNA during Selective Gene Transcription".

10. Saha S, Ansari AZ, Jarell KA, and Ptashne M, "RNA Sequences that Work as Transcriptional Activating Regions".

11. Frenster JH, "Ultrastructural Probes of Active DNA Sites, and the RNA Activators of DNA".