Submitted July 17, 1997.
Dynamics of telomere sequences are considered in normal and immortalized cells. Immortalized cells are suggested to be derived mainly from a special subpopulation of ontogenetic reserve cells. Their epigenetic program consists of autoregeneration during external stimulus for genome reorganization and corresponding appearance of genetic variants of cells. It is suggested that cells surviving the crisis stage contain a special signal sequence integrated into telomere DNA. Its elimination during shortening of DNA telomere sequences in dividing ontogenetic reserve cells is a signal for the cooperative transition of chromatin into a new steady state that corresponds to the epigenotype of immortalized cells. Localization of telomere DNA sequences in intrachromosomal "hot points" reflects phylogenetic rearrangement of the genome.
KEY WORDS: telomere, telomere DNA, ontogenetic reserve cells, chromatin phenotype, cooperative transitions, DNA loops, nuclear matrix, epigenetic state, chromosomal "hot points".
Data on the fate of telomere sequences in transformed cells are of principal importance. During the complex process of transformation the phenomenon of cell ageing is overcome and at a certain stage of this process cells become immortalized. However, the latter is occurs for only some of the cells. During transcription most cells live longer than their normal precursors, but gradually they also stop dividing and die. This is the so-called crisis phase. However, a small number of cells (for example, with a frequency of 10-7 for human fibroblasts infected with SV40 ) becomes immortalized, i.e., they overcome both replicative ageing and the crisis. This raises at least two problems: 1) Do initially normal cells overcoming crisis differ from other cells of the population? 2) Why is genome instability sharply increased in the process of transformation whereas in cells that overcome the crisis the number of chromosomal aberrations is stabilized?
Finally, there is a third problem. Telomere DNA sequences were shown to be localized not only in telomeres; they are distributed over other loci of vertebrate chromosomes [17, 20]. It is suggested that their appearance in non-telomere regions of chromosomes is due to chromosome fusion with subsequent amplification. These intrachromosomal repeats of DNA are sensitive to chromosome breaks and recombinations, i.e., they represent "hot points" in chromosomes during formation of chromosome aberrations [17, 21, 22]. It is interesting that DNA sequence we have isolated produce a hybridization signal not only near the telomere but also in a number of intrachromosome regions appearing as a result of chromosome fusions . Thus, we propose the third problem: 3) Why are telomere repeats often "hot points" by which chromosome aberrations are generated?
In this paper we do not present a review of "telomere biology". The references used are related only to those studies that can by useful to elucidate a number of questions. Moreover, some publications containing contradictions from the enumerated problems are deliberately left out of consideration. Thus, we raise a number of questions and in subsequent chapters we introduce our view concerning their answers.
Transition through Crisis to Cells with Stabilized DNA Telomere Lengths and the Question of Ontogenetic Reserve Cells
The dynamics of changes of DNA telomere sequences during cell divisions and telomerase activity have been studied in normal and SV40 infected cells of human embryonal kidney . In accordance with Olovnikovs prediction and data from other laboratories  the length of the telomere repeat gradually decreased during cells divisions, and the rate of this decrease is about 65 pairs of bases per division. This process also takes place in a population infected by SV40. It is characterized by a longer life-time (with an additional number of divisions) and cells become non-viable at the crisis point. We designate this point as the CP (crisis point). During transition through the CP the length of telomere repeats decreased from ~17 to ~1.5 thousand pairs of bases. Telomerase activity is absent both in dividing cells of the normal population and in transformed cells terminating their life at the CP. However, a small number of cells pass the CP and become immortalized. In these cells the length of decreased telomeres stabilize and telomerase activity appears. It was shown  that before CP a number of chromosome aberrations dramatically increased. However, after CP their number stabilized and even decreased.
What is the mechanism of the process promoting transition of some cells through CP that leads to their immortalization and is phenomenologically related to a decrease in length and then to stabilization of this parameter of DNA telomere sequences? In  it is suggested that this process is due to a shift of a heterochromatin block of telomere repeats during the decrease of their number into the region of a chromosome near telomeres. This is accompanied by heterochromatization of regulatory and/or structural genes located in the region near telomeres resulting in changes in their functional activity. However, we suggest that this situation can occur only during cell ageing. It is not necessarily related to the nearest to telomere sequences. For example, it is known that ribosomal genes of studied cells  are in the telomeric region, i.e., significantly closer to telomeres than to the centromere of chromosomes. As a result of shortening of telomere sequences the effect of a chromosome field  and also a sequence of heterochromatization  of active copies of ribosome genes will not function. This must result in cell ageing and finally in cell death. This mechanism does not contradict data on the activation and/or repression of a number of oncogenes in transformed cells, proteins of the cell cycle, etc. These processes are a specific stress for cells which leads to many of the mentioned molecular events.
However, this probable mechanism does not answer the question: why the transition throughout the CP requires shortening of the lengths of telomere sequences to certain critical value and why the phenomenon of immortalization takes place only in a small number of the total pool of either normal or transformed cells.
Let us begin the consideration of this problem from the last question. It is suggested that cells overcoming CP do not differ in their origin from cells lacking the possibility of overcoming CP. In other words, the CP transition is a stochastic event in a population of ageing cells . This conclusion is based on data reported in . The authors of this study demonstrated that diploid human fibroblasts can be divided into two groups by their proliferative potential. Fibroblasts of the first group can undergo a limited number of divisions. Cells of the second group undergo a more significant number of divisions. The probability of cells to be in the state with minor proliferative potential increases with the number of cell divisions. Finally after a relatively large number of divisions the main portion of cells will be in the state with low proliferative potential. In this case only a small number of cells will be able to perform subsequent divisions. It was also shown that the ability of cells to acquire different numbers of subsequent divisions is realized within one mitosis . During division of the parent cell each of the filial cells is divided until CP 10-20 and a few times, respectively. Only in 6% of pairs from all investigated dividing cells one of the filial cells passes through more than 20 divisions. The authors of  stress that now it is difficult to ascribe the above mentioned situation directly to telomere shortening. At least this requires the direct demonstration that a small number of cells passing through a large number of divisions had an initially longer telomere sequence.
There is another way to explain cell peculiarities allowing the overcoming of CP. In fact, it is known that the diploid genome of eucaryotic cells possess a significantly higher reliability than those of other karyotaxons [29, 30]. However, increased reliability of the genome implies minimization of mutational variation and, consequently, minimization of evolutionary possibilities. This could lead to an evolutionary dead-end because as a result of the pressure of external factors such "reliable genes" would be eliminated by chance. Some authors have tried to clarify this problem [31, 32]. They suggest that various mutations in relatively reliable genetic systems induce errors in the functioning of the system responsible for elimination of DNA damage. This is the SOS-repair. These authors suggest that mutations appearing during the functioning of the system of SOS-repair are important for evolutionary events and are preconditions for the appearance of tumor cells. At the same time mutational events are a stochastic process and therefore it is difficult to follow logically the relationship between mutations induced by a system of SOS-repair and the ability of a small cell population to overcome CP.
We believe that cells overcoming CP had initially different capacities than the larger number of cells of this population. The existence of such cells solves the contradiction between genome reliability and a need for some genetic variation. Our concept is the following: among a given population of cells heterogeneous in some characters (including proliferative potential) a small specialized subpopulation of cells that reacts to low doses of various stimuli by epigenetically programmed reactions always exist. This reaction consists in autoinduction in these cells of karyotype abnormalities and/or other mutational events. This leads to formation of new genetic variants and to selection from them of the variants that are most adapted to the altered or expected changes of the environment. Such a small subpopulation of cells we denominate as cells of evolutionary reserve [33-35]. However, it is more precise to call them cells of ontogenetic reserve . Some variants appearing may represent cells that are in the process of neoplastic transformation. This concept is given in detail in other publications [33, 35]. However, the subsequent analysis requires the following.
Cell chromatin, the substrate of chromosomes, represents a phase-separated system of molecules (DNA, proteins, etc.) that are in a programmed medium in certain spatial, physical, and chemical interrelationships. Owing to conformational capacities of macromolecules and weak interactions between certain elements of this supramolecular system it can be arranged by various modes that depend on parameters of the external medium. Consequently, both external factors, and the internal composition of chromatin will determine its various of condition in the "space of possibilities". Thus, chromatin represents a wider notion than chromosomes. The latter are a certain concrete form of its existence in a given cell which is in a certain physiological state. In other words, chromatin is a system possessing higher degrees of freedom than it is realized in a given cell. So, chromatin is able to choose one condition from a number of possible ones. All this allows consideration of the idea as the phenotype of chromatin because "phenotype can be considered as a possibility of choice from several possible ways of realization of information transported from chromosomal DNA" . The notion "phenotype of chromatin or nucleus" is very important because in contrast to other cellular systems, chromatin structure is realized and changed during transcription and translation of the information encoded in its element, DNA. Not only transcribed but all DNA of a cell is important because it participates in spatial organization of the considered system. Time is the age of the system. The accumulated time is the rate of structural-functional organization of the system, i.e., structures formed during interaction with products appearing as the result of transcription and translation of the information encoded in it and rearranging it. This leads to transcription of a new information, etc., until achievement of some new steady state condition appears. Thus, steady state of chromatin determine a character of realized information and, consequently, the entire subsequent fate of the cell. According to Olovnikovs classification such a steady state of chromatin is fatomere . There is some evidence to suggest  that at least each differentiated cell possesses its distinct steady state of chromatin. Chromatin structure revealed cytologically is considered as a "portrait of the tissue" . In this case the number of steady states of chromatin is not less than the number of differentiated cells in eukaryotes. This number contains a few hundreds of states . However, even in a given cell chromatin can be in a few discrete conditions. For hepatocytes their numbers are 4 . In addition, some genetic pathologies have their own characteristic chromatin phenotype [43-46]. This implies that the number of normal chromatin states does not limit all the possible spectrum of states. Analysis revealed  that differences in epigenetic programs of various differentiated cells are stipulated by the transition in the functioning of only a small number of the loops on each chromosome. But macroscopic differences ("portrait of the tissue") appear in chromatin organization in nuclei of these cells. This suggests that transitions between steady states of chromatin represent a cooperative process influencing spatial organization of the genome in a cell. One of the possible models for such a transition was presented in . Thus, the existence of not forbidden but not realized states is possible under normal conditions. The phenotype of nuclei of ontogenetic reserve cells is such a new condition. Induced autogeneration of various genetic impairments is a new program for these cells. Typical for the phenotype of these cells, the repression of DNA-repair systems and a significant decrease in the activity of antioxidant enzymes, superoxide dismutase, catalase, and glutathione reductase can be the main mechanism for such events. It is known that these enzymes utilize active oxygen radicals and hydroperoxides that are formed during cell metabolism and which induce various types of DNA damage. In forming genetic variants of these cells the activity of these enzymes of DNA defense is restored.
Let us consider a possible mechanism allowing cells of ontogenetic reserve to realize a program of autogeneration of molecular impairments in their genome at critical length of DNA telomere sequence and simultaneously to overcome CP. It is known that in normal cells DNA telomere sequences are associated with a molecular matrix [16, 17, 48]. There are two types of interaction with the matrix . The first is related to packaging and condensation of chromatin and is relatively stable for each type of cell. The second represents transient interaction of DNA with the matrix that depends on transcriptional and replicative activity of a given DNA sequence. In the later case, the interaction with the matrix is required for activation of these processes. Dissociation of a loop from the matrix corresponds to inactivation of genes localized in it [50, 51]. It is suggested that the interaction of heterochromatinized chromatin containing DNA telomere sequences occurs via the first type . Near-telomere sequences of DNA contain structural and regulatory genes. Their inactivation with corresponding consequences for the cell can be achieved via their heterochromatization during telomere shortening  or via dissociation of a loop from the matrix. However, only a significant decrease of telomere length does not lead to the transit of the major part of the cells through CP . Thus, eliminated telomere repeats are probably not signals for the cell transit through CP. In this case, shortening of telomere sequences leads to gradual cell ageing. Perhaps, this is related (as we have already mentioned above) with repression of some copies of ribosome genes. Transition via CP is a cooperative process and is occurs by a jump. So, if this transit is stipulated by telomere shortening, we ought to suggest another mechanism which is considered below.
We suggest that the signal for such a transition must be another DNA sequence integrated into a telomere and being a region of fastening of some near-telomere sequences to the matrix. Let us call this DNA fragment a CP sequence. The existence of such a sequence and its elimination during telomere shortening is a signal (unfastening) of the cooperative transition to a new steady state, the phenotype of immortalized cells. We suggest that such a signal exists only in cells of ontogenetic reserve and, perhaps, in sex cells. However, their epigenotype programs cause autogeneration of genome impairments (see above). Thus, CP the transition is accompanied by genome instability and a dramatic decrease of chromosome aberrations [17, 24]. Only the transition of the CP point and stabilization of a new phenotype lead to activation (expression) of telomerase, DNA repair, and antioxidant systems. This is accompanied by a decrease of chromosome abnormalities , stabilization of telomere length . The possible existence of CP-sequences has indirect confirmation. In a number of cells the telomere sequences are localized not only in the telomere region of the chromosomes . They are also distributed in other chromosome loci . Among such telomere repeats distributed over the chromosome non-telomere repeats were also found. For example, it was shown that about 270 pairs of bases of a seldomly copied DNA are localized between inverted repeats of telomere sequences located in near-telomere region of mouse chromosome 2 . The existence of a unique or seldomly copied DNA between inverted telomere sequence was also demonstrated for human genome DNA . Since in some cell types overcoming CP the length of telomere sequence is about 1500 pairs  and we suggest that the CP-repeat is integrated into a telomere repeat at the same distance from the near-telomere chromosome region.
Thus, we suggest that in a small subpopulation of cells of the ontogenetic reserve in contrast to other cell types (sex?) the DNA telomere sequences contain some additional CP-sequence. This is a region that fastens a loop lying near theee telomere where a number of near-telomere DNA sequences are localized. Elimination of the CP sequence during telomere shortening leads to loop unfastening and this is a signal for transition of chromatin into a new steady state characteristic for immortalized cells.
DNA Telomere Sequences and "Hot Points" in Chromosomes
It is known that the number of chromosome aberrations dramatically increases in cells just before events of transition through CP. In cells that overcome CP, the number of chromosome aberrations gradually stabilizes and telomerase activity appears . As mentioned above, telomere sequences integrated into internal chromosome regions are "hot points" involved into expression of many chromosome aberrations . For understanding of this phenomenon it is relevant to come back to the cells of the ontogenetic reserve. Changes of the genome in these cells during their development from "common" ("embryonal") state to the stage of autogeneration of genetic impairments in them can be considered as ontogenesis of the genome. In this case loci of most the frequent impairments in the genome can reflect its phylogenesis. In other words, these loci are subjected to (or were the places of) rearrangements of the genome during its evolution. For example, localization of telomere sequences in the internal sites of chromosomes is often characterized as places of their fusions in more primitive cells. Thus, not only evolutionaary changes of the genome but the process of mutagenesis are reflected as "hot points" in cells of the ontogenetic reserve (and in other cells as well). From these positions it is worthwhile to investigate the "evolutionary" nature of other displacing sequences of DNA . We agree with theee authors of  who suggest that higher stability of telomere sequences to mutational changes is the a function of their compartmentatization in the cell nucleus. Apparently, this is also correct for other displaced sequences of DNA. Genome reorganization that occurs during autogeneration of these events in cells of ontogenetic reserve via "hot points" is probably the optimal (the cell "remembers" these points) for manifestation of new genetic variants, some of which are ancestors of immortalized clones. This "optimality" is due to events induced in "hot points" which must lead also to the recapitulation of genetic variants which have passed selection earlier. However, these "ancestor" variants presently contain new information because they were enriched with new mutations. Hence, these variants possess new possibilities for surviving. In addition, it is known that in telomeric regions of chromosomes, chromatin is in a relaxed conformation available for endonucleases . These data confirm the suggestion that intercalary repeating telomeric sequences of DNA increase the probability of interchromosome interactions resulting in recombinations and exchanges [21, 52]. Finally, it has been shown that alpha-particle-induced multiple damages are concentrated in a relatively small area of the chromosome arms . The latter explains to some extent a high frequency of interactions between damage and its possible passage to cells of the following generations , i.e., of new genetic variants.
Cells of the ontogenetic reserve manifest putatively mutatory phenotype. Therefore induction in them of the program of autogeneration of genomic transformations, including those in "hot points", must result in the manifestation of new genetic variants with rather high frequency. This condition is necessary for the fixation of "favorable" variants in the cell population  in which ontogenetic reserve cells present just a small subpopulation.
The study was performed with partial financial support from the State Scientific Technical Program "Human Genome" (grant No. 69/97).
1.Olovnikov, A. M. (1971) Dokl. Akad. Nauk
SSSR, 201, 1496-1499.
2.Olovnikov, A. M. (1973) J. Theor. Biol., 41, 181-190.
3.Hayflick, L., and Moorhead, P. S. (1961) Exp. Cell. Res., 25, 565-621.
4.Harley, C. B. (1991) Mutat. Res., 256, 271-282.
5.Greider, C. W., and Blackburn, E. H. (1985) Cell, 43, 405-413.
6.Greider, C. W., and Blackburn, E. H. (1987) Cell, 51, 887-898.
7.Allshire, R. C., Gisten, J. R., Cross, S. H., Cranston, G., Rout, D., Sugavara, N., Szostak, J. W., and Hastie, N. D. (1988) Nature, 332, 656-659.
8.Morin, G. B., (1989) Cells, 59, 521-529.
9.Cooke, H. J., Brown, W. R. A., and Rappold, G. A. (1985) Nature, 317, 687-692.
10.De Lange, T., Shiue, L., Myers, R. M., Cox, D. R., Naylor, S. L., Rillery, A. M., and Varmus, H. E. (1990) Mol. Cell. Biol., 10, 518-527.
11.Hastie, N. D., Dempster, M., Dunlop, M. G., Thompson, A. M., Green, D. K., and Allshire, R. C. (1990) Nature, 346, 866-868.
12.McClintock, B. (1941) Genetics, 26, 234-282.
13.Blackburn, E. H., and Szostak, J. W. (1984) Ann. Rev. Biochem., 53, 163-194.
14.Zakian, V. A. (1989) Ann. Rev. Genet., 23, 579-604.
15.Blackburn, E. H. (1991) Nature, 350, 569-573.
16.De Lange, T. (1992) EMBO J., 11, 717-724.
17.Balajee, A. S., Dominguez, I., Bohr, V. A., and Natarajan, A. T. (1996) Mutat. Res., 372, 163-172.
18.Raghuramann, M. K., Brewer, B. J., and Fangman, W. L. (1997) Science, 276, 806-809.
19.Wright, W. E., and Shay, J. W. (1992) TIG, 8, 193-197.
20.Meyne, J., Baker, R. J., Hobart, H. H., Hsu, T. C., Ryder, O. A., Ward, O. G., Wiley, J. E., Wurster-Hill, D. H., Yates, T. L., and Moyzis, R. K. (1990) Chromosoma, 99, 3-10.
21.Hastie, N. D., and Allshire, R. C. (1989) Trends Genet., 5, 326-331.
22.Alvarez, L., Evans, J. W., Wilks, R., Lucas, J. N., Brown, J. M., and Giaccia, A. J. (1993) Genes. Chromosomes. Cancer, 8, 8-14.
23.Karpukhin, A. V., Bogush, A. I., Salimov, A. G., Veiko, N. N., Nemtsova, M. V., Zakharev, V. M., Zaletaev, D. B., and Spitkovsky, D. M. (1993) Dokl. Ros. Akad. Nauk, 329, 365-368.
24.Counter, C. M., Avilion, A. A., Le Feuvre, C. E., Stewart, N. G., Greider, C. W., Harley, C. B., and Bacchetti, S. (1992) EMBO J., 11, 1921-1928.
25.Lima-de-Faria, A. (1976) Hereditas, 83, 1-22.
26.Lima-de-Faria, A. (1976) Hereditas, 84, 19-34.
27.Smith, J. R., and Whitney, R. G. (1980), Science, 207, 82-84.
28.Jones, R. B., Whitney, R. G., and Smith, J. R. (1985) Mech. Ageing Dev., 29, 143-149.
29.Karogodin, V. I. (1982) Radiobiologiya, 22, 147-154.
30.Sarapul'tsev, B. I., and Geraskin, S. A. (1993) Genetic Bases of Radioresistance and Evolution [in Ruissian], Energoatomizdat, Moscow, p. 209.
31.Edhols, H. (1981) Cell, 25, 1-2.
32.Edhols, H., and Goodman, M. (1990) Mutat. Res., 236, 301-311.
33.Spitkovsky, D. M. (1992) Radiobiologiya, 32, 382-400.
34.Spitkovsky, D. M. (1992) Abst. Int. Conf. Low Dose Irradiation and Biol. Defence Mechanisms, Kyoto, PD-3, p. 140.
35.Spitkovsky, D. M., and Talyzina, T. A. (1995) Rad. Protect. Dosimetry, 62, 23-26.
36.Spitkovsky, D. M. (1997) Proc. Third Congr. on Radiation Studies, Moscow, in press.
37.Woddingtop, K. H. (1970) On the Way to Theoretical Biology [Russian translation], Mir, Moscow, pp. 11-38.
38.Olovnikov, A. M. (1996) Biochemistry (Moscow), 61, 1948-1970 (Russ.).
39.Spitkovsky, D. M. (1979) in Theoretical Problems of Medical Genetics (Zakharov, A. F., ed.) Meditsina, Moscow, pp. 52-73.
40.Prokofeva-Belgovskaya, A. A. (1992) Mol. Biol. (Moscow), 16, 771-775.
41.Konyukhov, B. V., Kupriyanov, S. D., and Isambekov, B. S. (1988) Uspekhi Sovr. Genet., 15, 106-129.
42.Nemirovsky, L. E., Vakhtel, H. M., Solov'eva, I. G., and Kormiletz, M. V. (1987) Dokl. Akad. Nauk SSSR, 292, 468-472.
43.Passarge, E. (1972) Hum. Genet., 16, 151-157.
44.Green, M. C., Schultz, L. D., and Nedsi, L. A. (1975) Transplantation, 20, 172-175.
45.Karpukhin, A. V., Fedorova, K. N., and Spitkovsky, D. M. (1978) Byull. Eksp. Biol. Med., No. 1, 471-474.
46.Puvion-Difiller, F., and Saransin, A. (1989) Mutat. Res., 219, 57-61.
47.Andreev, S. G., and Spitkovsky, D. M. (1983) Dokl. Akad. Nauk SSSR, 269, 1500-1503.
48.De Lara, J., Wydner, K. L., Huband, K. M., and Ward, W. S. (1993) J. Cell Biochem., 53, 213-221.
49.Bodnar, J. W. (1988) J. Theor. Biol., 132, 479-507.
50.Andreev, S. G., Spitkovsky, D. M., and Talyzina, T. A. (1987) Proc. Fifth All-Union Meet. on Microdosimetry, Vol. 2, MIFI, Moscow, pp. 4-60.
51.Jackson, D. A., Dolle, A., Robertson, G., and Cook, P. R. (1992) Cell Biol. Int. Rep., 16, 687-696.
52.Bouffler, S., Silver, A., Papworth, D., Coates, J., and Cox, R. (1993) Genes. Chromosomes. Cancer, 6, 98-106.
53.Weber, B., Collins, C., Robins, C., Magenis, R. E., Delaney, A. D., Gray, J. W., and Havden, M. R. (1990) Nucleic Acids Res., 18, 3353-3361.
54.Karpukhin, A. V., Bogush, A. I., Zakharev, V. M., Salimov, A. G., Beiko, N. N., Spitkovsky, D. M., Fedorova, L. I., Zelenin, A. V., Bannikov, V. M., Domninsky, D. M., Zabarovsky, E. R., and Kiselev, L. L. (1996) Mol. Biol. (Moscow), 6, 1339-1347.
55.Ferraro, M., Predazzi, V., and Prantera, G. (1993) Chromosoma, 102, 712-717.
56.Brenner, D. J., and Sachs, R. K. (1994) Radiat. Res., 140, 134-142.
57.Griffin, C. S., Marsden, S. J., Steven, D. L., Simpson, P., and Savage, J. R. K. (1995) J. Radiat. Res., 67, 431-439.
58.Taddel, F., Radman, M., Maynard-Smith, J., Toupance, B., Gouyon, P. H., and Godelle, B. (1997) Nature, 387, 700-702.