WO1995013382A1 - Therapy and diagnosis of conditions related to telomere length and/or telomerase activity - Google Patents

Abstract

Method and compositions are provided for the determination of telomere length and telomerase activity, as well as the ability to increase or decrease telomerase activity in the treatment of proliferative diseases. Particularly, primers are elongated under conditions which minimize interference from other genomic sequences, so as to obtain accurate determinations of telomeric length or telomerase activity. In addition, compositions are provided for intracellular inhibition of telomerase activity and means are shown for slowing or reversing the loss of telomeric repeats in aging cells.

Description

DESCRIPTION

THERAPY AND DIAGNOSIS OF CONDITIONS RELATED TO TELOMERE LENGTH AND/OR TELOMERASE ACTIVITY

This invention relates to methods for therapy and diagnosis of cellular senescence and immortalization. Background of the Invention The following is a general description of art relevant to the present invention. None is admitted to be prior art to the invention. Generally, this art relates to observations relating to cellular senescence, and theories or hypotheses which explain such aging and the mechanisms by which cells escape senescence and immortalize.

Normal human somatic cells (e.g. , fibroblasts, endothelial, and epithelial cells) display a finite replicative capacity of 50-100 population doublings characterized by a cessation of proliferation in spite of the presence of adequate growth factors. This cessation of replication in vi tro is variously referred to as cellular senescence or cellular aging, See, Goldstein, 249 Science 1129, 1990; Hayflick and Moorehead, 25 Exp. Cell Res. 585, 1961; Hayflick, ibid.. 37:614, 1985; Ohno, 11 Mech. Aging Dev. 179, 1979; Ham and McKeehan, (1979) "Media and Growth Requirements", .B. Jacoby and I.M. Pastan (eds) , in: Methods in Enzymology, Academic Press, NY, 58:44-93. The replicative life span of cells is inversely proportional to the in vivo age of the donor (Martin et al. , 23 Lab. Invest. 86, 1979; Goldstein et al., 64 Proc. Natl. Acad. Sci. USA 155, 1969; and Schneider and Mitsui, ibid. , 73:3584, 1976) , therefore cellular senescence is suggested to play an important role in aging in vivo. Cellular immortalization (the acquisition of unlimited replicative capacity) may be thought of as an abnormal escape from cellular senescence, Shay et al. , 196 Exp. Cell Res. 33, 1991. Normal human somatic cells appear to be mortal, i.e. , have finite replicative potential. In contrast, the germ line and malignant tumor cells are immortal (have indefinite proliferative potential) . Human cells cultured in vitro appear to require the aid of transforming viral oncoproteins to become immortal and even then the frequency of immortalization is IO^"6 to IO^"7. Shay and Wright, 184 Exp. Cell Res. 109, 1989. A variety of hypotheses have been advanced over the years to explain the causes of cellular senescence. While examples of such hypotheses are provided below, there appears to be no consensus or universally accepted hypothesis .

For example, the free radical theory of aging suggests that free radical-mediated damage to DNA and other macromolecules is causative in critical loss of cell function (Harman, 11 J. Gerontol. 298, 1956; Harman, 16 J. Gerontol. 247, 1961) . Harman says (Harman, 78 Proc. Natl. Acad. Sci. 7124, 1981) "aging is largely due to free radical reaction damage..."

Waste-product accumulation theories propose that the progressive accumulation of pigmented inclusion bodies (frequently referred to as lipofuscin) in aging cells gradually interferes with normal cell function (Strehler, 1 Adv. Geront . Res. 343, 1964; Bourne, 40 Prog. Brain Res . 187, 1973; Hayflick, 20 Exp. Gerontol. 145, 1985) .

The somatic mutation theories propose that the progressive accumulation of genetic damage to somatic cells by radiation and other means impairs cell function and that without the genetic recombination that occurs, for instance, during meiosis in the germ line cells, somatic cells lack the ability to proliferate indefinitely

(Burnet, "Intrinsic Mutagenesis - A Genetic Approach to

Aging", Wile, NY, 1976; Hayflick, 27 Exp. Gerontol . 363, 1992) . Theories concerning genetically programmed senescence suggest that the expression of senescent- specific genes actively inhibit cell proliferation (Martin et al., 74 Am. J. Pathol. 137, 1974; Goldstein, 249

Science 1129, 1990) .

Smith and Whitney, 207 Science 82, 1980, discuss a mechanism for cellular aging and state that their data is:

"compatible with the process of genetically controlled terminal differentiation.. . . The gradual decrease in proliferation potential would also be compatible with a continuous build up of damage or errors, a process that has been theorized. However, the wide variability in doubling potentials, especially in mitotic pairs, suggests an unequalled partitioning of damage or errors at division."

Shay et al., 27 Experimental Gerontology 477, 1992, and 196 Exp. Cell Res. 33, 1991 describe a two- stage model for human cell mortality to explain the ability of Simian Virus 40 T-antigen to immortalize human cells. The mortality stage 1 mechanism (Ml) is the target of certain tumor virus proteins, and an independent mortality stage 2 mechanism (M2) produces crisis and prevents these tumor viruses from directly immortalizing human cells. The authors utilized T- antigen driven by a mouse mammary tumor virus promoter to cause reversible immortalization of cells. The Simian Virus 40 T-antigen is said to extend the replicative life span of human fibroblast by an additional 40-60%. The authors postulate that the Ml mechanism is overcome by T-antigen binding to various cellular proteins, or inducing new activities to repress the Ml mortality mechanism. The M2 mechanism then causes cessation of proliferation, even though the Ml mechanism is blocked. Immortality is achieved only when the M2 mortality mechanism is also disrupted.

It has also been proposed that the finite replicative capacity of cells may reflect the work of a "clock" linked to DNA synthesis in the telomere (end part) of the chromosomes. Olovnikov, 41 J. Theoretical Biology 181, 1973, describes the theory of marginotomy to explain the limitations of cell doubling potential in somatic cells. He states that an:

"informative oligonucleotide, built into DNA after a telogene and controlling synthesis of a repressor of differentiation, might serve as a means of counting mitosis performed in the course of morphogenesis. Marginotomic elimination of such an oligonucleotide would present an appropriate signal for the beginning of further differentiation. Lengthening of the telogene would increase the number of possible mitoses in differentiation."

Harley et al. , 345 Nature 458, 1990, state that the amount and length of telomeric DNA in human fibroblasts decreases as a function of serial passage during aging in vi tro, and possibly in vivo, but do not know whether this loss of DNA has a causal role in senescence. They also state: "Tumour cells are also characterized by shortened telomeres and increased frequency of aneuploidy, including telomeric associations. If loss of telomeric DNA ultimately causes cell-cycle arrest in normal cells, the final steps in this process may be blocked in immortalized cells. Whereas normal cells with relatively long telomeres and a senescent phenotype may contain little or no telomerase activity, tumour cells with short telomeres may have significant telomerase activity. Telomerase may therefore be an effective target for anti-tumour drugs.

There are a number of possible mechanisms for loss of telomeric DNA during ageing, including incomplete replication, degradation of termini (specific or nonspecific) , and unequal recombination coupled to selection of cells with shorter telomeres. Two features of our data are relevant to this question. First, the decrease in mean telomere length is about 50 bp per mean population doubling and, second, the distribution does not change substantially with growth state or cell arrest. These data are most easily explained by incomplete copying of the template strands at their 3' termini. But the absence of detailed information about the mode of replication or degree of recombination at telomeres means that none of these mechanisms can be ruled out. Further research is required to determine the mechanism of telomere shortening in human fibroblasts and its significance to cellular senescence." [Citations omitted.]

Hastie et al. , 346 Nature 866, 1990, while discussing colon tumor cells, state that: "[T]here is a reduction in the length of telomere repeat arrays relative to the normal colonic mucosa from the same patient. . . .

Firm figures are not available, but it is likely that the tissues of a developed fetus result from 20-50 cell divisions, whereas several hundred or thousands of divisions have produced the colonic mucosa and blood cells of 60-year old individuals. Thus the degree of telomere reduction is more or less proportional to the number of cell divisions. It has been shown that the ends of Drosophila chromosomes without normal telomeres reduce in size by _4 base pairs (bp) per cell division and that the ends of yeast chromosomes reduce by a similar degree in a mutant presumed to lack telomerase function. If we assume the same rate of reduction is occurring during somatic division in human tissues, then a reduction in TRA by 14 kb would mean that 3,500 ancestral cell divisions lead to the production of cells in the blood of a 60-year old individual; using estimates of sperm telomere length found elsewhere we obtain a value of 1,000-2,000. These values compare favourably with those postulated for mouse blood cells. Thus, we propose that telomerase is indeed lacking in somatic tissues. In this regard it is of interest to note that in maize, broken chromosomes are only healed in sporophytic (zygotic) tissues and not in endosperm (terminally differentiated) , suggesting that telomerase activity is lacking in the differentiated tissues." [Citations omitted.]

The authors propose that in some tumors telomerase is reactivated, as proposed for HeLa cells in culture, which are known to contain telomerase activity. But, they state:

"One alternative explanation for our observations is that in tumours the cells with shorter telomeres have a growth advantage over those with larger telomeres, a situation described for vegetative cells of tetrahymena. " [Citations omitted.]

Harley, 256 Mutation Research 271, 1991, discusses observations allegedly showing that telomeres of human somatic cells act as a mitotic clock shortening with age both in vi tro and in vivo in a replication dependent manner. He states:

"Telomerase activation may be a late, obligate event in immortalization since many transformed cells and tumour tissues have critically short telomeres.

Thus, telomere length and telomerase activity appear to be markers of the replicative history and proliferative potential of cells; the intriguing possibility remains that telomere loss is a genetic time bomb and hence causally involved in cell senescence and immortalization. Despite apparently stable telomere length in various tumour tissues or transformed cell lines, this length

5 was usually found to be shorter than

„ those of the tissue of origin.

These data suggest that telomerase becomes activated as a late event in cell transformation, and that cells

10 could be viable (albeit genetically unstable) with short telomeres stably maintained by telomerase. If telomerase was constitutively present in a small fraction of

15 normal cells, and these were the ones which survived crisis or became transformed, we would expect to find a greater frequency of transformed cells with long telomeres . "

20 [Citations omitted.]

He proposes a hypothesis for human cell aging and transformation as " [a] semi-quantitative model in which telomeres and telomerase play a causal role in cell 25 senescence and cancer" and proposes a model for this hypothesis .

De Lange et al. , 10 Molecular and Cellular Biology 518, 1990, generally discuss the structure of human chromosome ends or telomeres. They state:

30 "we do not know whether telomere reduction is strictly coupled to cellular proliferation. If the diminution results from incomplete replication of the telomere, such a

35 coupling would be expected; however, other mechanisms, such as exonucleolytic degradation, may operate independent of cell division. In any event, it is clear

40 that the maintenance of telomeres is impaired in somatic cells . An obvious candidate activity that may

, be reduced or lacking is telomerase.

A human telomerase activity that can

45 add TTAGGG repeats to G-rich primers has recently been identified (G. Morin, personal communication) . Interestingly, the activity was demonstrated in extracts of HeLa

50 cells, which we found to have exceptionally long telomeres . Other cell types have not been tested yet, but such experiments could now establish whether telomerase activity is (in part) responsible for the dynamics of human chromosome ends . "

Kipling and Cooke, 347 Nature 400, 1990, indicate that mice have large telomeres and discusses this length in relationship to human telomeres . In regard to mice telomers, they state:

"Whether long telomeres are a result of selection or simply a neutral change is not clear. Their size seems largely unchanged on passage to subsequent generations, as well as through somatic cell division, so it is unlikely that the extra length is a defence against rapid loss of sequence. Nor are mouse telomeres significantly reduced in size during the animal's lifespan; a 17-month- old individual still showed normal size distribution of fragments characteristic of its strain (data not shown) . This, and the much longer telomeres of this short-lived species, suggests that telomere shortening is unlikely to have any causal role in ageing in vivo, in contrast to some recent speculations . The shortening of human telomeres during ageing in vivo may instead indicate that telomere maintenance is another metabolic process that senescent cells are unable to perform as efficiently. "

D'Mello and Jazwinski, 173 J. Bacteriology

6709, 1991, State:

"We propose that during the life span of an organism, telomere shortening does not play a role in the normal aging process. However, mutations or epigenetic changes that affect the activity of the telomerase, like any other genetic change, might affect the life span of the individual in which they occur. In summary, the telomere shortening with age observed in human diploid fibroblasts may not be a universal phenomenon. Further studies are required to examine telomere length and telomerase activity not only in different cell types as they age but also in the same cell type in different organisms with differing life spans. This would indicate whether telomere shortening plays a causal role in the senescence of a particular cell type or organism. "

Hiyama et al., 83 Jpn. J. Cancer Res. 159, 1992, provide findings that "suggest that the reduction of telomeric repeats is related to the proliferative activity of neuroblastoma cells and seems to be a useful indicator of the aggressiveness of neuroblastoma.. . . Although we do not know the mechanism of the reduction and the elongation of telomeric repeats in neuroblastoma, we can at least say that the length of telomeric repeats may be related to the progression and/or regression of neuroblastoma."

Counter et al., 11 EMBO J. 1921, 1992, state "loss of telomeric DNA during cell proliferation may play a role in ageing and cancer." They propose that the expression of telomerase is one of the events required for a cell to acquire immortality and note that:

This model may have direct relevance to tumourigenesiε in vivo . For example, the finite lifespan of partiallytransformed (pre-immortal) cells which lack telomerase might explain the frequent regression of tumours after limited growth in vivo . In bypassing the checkpoint representing normal replicative senescence, transformation may confer an additional 20-40 population doubling during which an additional <=→2 kbp of telomeric DNA is lost. Since 20-40 doubling (10⁶- 10¹² cells in a clonal population) potentially represents a wide range of tumour sizes, it is possible that many benign tumours may lack telomerase and naturally regress when telomeres become critically shortened. We predict that more aggressive, perhaps metastatic tumours would contain immortal cells which express telomerase. To test this hypothesis, we are currently attempting to detect telomerase in a variety of tumour tissues and to correlate activity with proliferative potential. Anti- telomerase drugs or mechanisms to repress telomerase expression could be effective agents against tumours which depend upon the enzyme for maintenance of telomeres and continued cell growth.

Levy et al. , 225 J. Mol. Biol. 951, 1992, state that: "Although it has not been proven that telomere loss contributes to senescence of multicellular organisms, several lines of evidence suggest a causal relationship may exist.

It is also possible that telomere loss with age is significant in humans, but not in mice." [Citations omitted. ]

Windle and McGuire, 33 Proceedings of the

American Association for Cancer Research 594, 1992, discuss the role of telomeres and state that:

"These and other telomere studies point in a new direction regarding therapeutic targets and strategies to combat cancer. If the cell can heal broken chromosomes preventing genomic disaster, then there may be a way to facilitate or artificially create this process. This could even provide a preventive means of stopping cancer which could be particularly applicable in high risk patients. The difference in telomere length in normal versus tumor cells also suggests a strategy where the loss of telomeres is accelerated. Those cells with the shortest telomeres, such as those of tumor metastasis would be the most susceptible. "

Goldstein, 249 Science 1129, 1990, discusses various theories of cellular senescence including that of attrition of telomeres . He states :

"However, such a mechanism is not easily reconciled with the dominance of senescent HDF over young HDF in fusion hybrids, particularly in short-term heterokaryons . One could again invoke the concept of dependence and the RAD9 gene example, such that complete loss of one or a few telomeres leads to the elaboration of a negative signal that prevents initiation of DNA synthesis, thereby mimicking the differentiated state. This idea, although speculative, would not only explain senescent replicative arrest but also the chromosomal aberrations observed in senescent HDS that would specifically ensue after loss of telomeres." [Citations omitted.]

The role of telomere loss in cancer is further discussed by Jankovic et al. and Hastie et al . , both at 350 Nature 1991, in which Jankovic indicates that telomere shortening is unlikely to significantly influence carcinogenesis in men and mice. Hastie et al . agree that if telomere reduction does indeed reflect cell turnover, this phenomenon is unlikely to play a role in pediatric tumors, and those of the central nervous system. Hastie et al . , however, feel "our most original and interesting conclusion was that telomere loss may reflect the number of cell division in a tissue history, constituting a type of clock."

Kipling and Cooke, 1 Human Molecular Genetics 3, 1992, state:

"It has been known for some years that telomeres in human germline cells (e.g. sperm) are longer than those in somatic tissue such as blood. One proposed explanation for this is the absence of telomere repeat addition (i.e. absence of telomerase activity) in somatic cells. If so, incomplete end replication would be expected to result in the progressive loss of terminal repeats as somatic cells undergo successive rounds of division. This is indeed what appears to happen in vivo for humans, with both blood and skin cells showing shorter telomeres with increasing donor age, and telomere loss may contribute to the chromosome aberrations typically seen in senescent cells. Senescence and the measurement of cellular time is an intriguingly complex subject and it will be interesting to see to what extent telomere shortening has a causal role . The large telomeres possessed by both young and old mice would seem to preclude a simple relationship between telomere loss and ageing, but more elaborate schemes cannot be ruled out." [Citations omitted.]

Greider, 12 BioEssays 363, 1990, provides a review of the relationship between telomeres, telomerase, and senescence. She indicates that telomerase contains an RNA component which provides a template for telomere repeat synthesis . She notes that an oligonucleotide "which is complementary to the RNA up to and including the CAACCCCAA sequence, competes with d(TTGGGG)n primers and inhibits telomerase in vitro" (citing Greider and Blackburn, 337 Nature 331,

1989) . She also describes experiments which she believes "provide direct evidence that telomerase is involved in telomere synthesis in vivo . " She goes on to state: "Telomeric restriction fragments in many transformed cell lines are much shorter than those in somatic cells. In addition, telomere length in tumor tissues is significantly shorter than in the adjacent non- tumor tissue. When transformed cell lines are passaged in vi tro there is no change in telomere length. Thus if untransformed cells lack the ability to maintain a telomere length equilibrium, most transformed cells appear to regain it and to reset the equilibrium telomere length to a size shorter than seen in most tissues in vivo . The simplest interpretation of these data is that enzymes, such as telomerase, involved in maintaining telomere length may be required for growth of transformed cells and not required ^• for normal somatic cell viability. This suggests that telomerase may be a good target for anti-tumor drugs." [Citations omitted.]

Blackburn, 350 Nature 569, 1991, discusses the potential for drug action at telomeres stating: "The G-rich strand of the telomere is the only essential chromosomal DNA sequence known to be synthesized by the copying of a separate RNA sequence. This unique mode of synthesis, and the special structure and behavior of telomeric DNA, suggest that telomere synthesis could be a target for selective drug action. Because telomerase activity seems to be essential for protozoans or yeast, but not apparently for mammalian somatic cells, I propose that telomerase should be explored as a target for drugs against eukaryotic pathogenic or parasitic microorganisms, such as parasitic protozoans or pathogenic yeasts. A drug that binds telomerase selectively, either through its reverse-transcriptase or DNA substrate-bindingproperties, should selectively act against prolonged maintenance of the dividing lower eukaryote, but not impair the mammalian host over the short term, because telomerase activity in its somatic cells may normally be low or absent. Obvious classes of drugs to investigate are those directed specifically against reverse transcriptases as opposed to other DNA or RNA polymeraseε, and drugs that would bind telomeric DNA itself . These could include drugs that selectively bind the G°G base- paired forms of the G-rich strand protrusions at the chromosome termini, or agents which stabilize an inappropriate G°G base-paired form, preventing it from adopting a structure necessary for proper function in vivo . Telomeres have been described as the Achilles heel of chromosomes: perhaps it is there that drug strategies should now be aimed." [Citations omitted.]

Lundblad and Blackburn, 73 Cell 347, 1993, discuss alternative pathways for maintainance of yeast telomers, and state that:

"...the work presented in this paper demonstrates that a defect in telomere replication need not result in the death of all cells in a population, suggesting that telomere loss and its relationship to mammalian cellular senescence may have to be examined further."

Other review articles concerning telomeres include Blackburn and Szostak, 53 Ann. Rev. Biochem.

163, 1984; Blackburn, 350 Nature 569, 1991; Greider, 67

Cell 645, 1991, and Moyzis 265 Scientific American 48, 1991. Relevant articles on various aspects of telomeres include Cooke and Smith, Cold Spring Harbor Symposia on Quantitative Biology Vol. LI, pp. 213-219; Morin, 59 Cell 521, 1989; Blackburn et al. , 31 Genome 553, 1989; Szostak, 337 Nature 303, 1989; Gall, 344 Nature 108, 1990; Henderson et al . , 29 Biochemistry 732, 1990; Gottschling et al. , 63 Cell 751, 1990; Harrington and Grieder, 353 Nature 451, 1991; Muller et al . , 67 Cell 815, 1991; Yu and Blackburn, 67 Cell 823, 1991; and Gray et al. , 67 Cell 807, 1991. Other articles or discussions of some relevance include Lundblad and Szostak, 57 Cell 633, 1989; and Yu et al. , 344 Nature 126, 1990.

Summary of the Invention This invention concerns methods for therapy and diagnosis of cellular senescence and immortalization utilizing techniques associated with control of telomere length and telomerase activity. Therapeutic strategies of this invention include reducing the rate or absolute amount of telomere repeat length loss or increasing the telomere repeat length during cell proliferation, thereby providing for the postponement of cellular senescence and reducing the level of chromosomal fusions and other chromosomal aberrations. In addition, inhibition of telomerase activity in vivo or in vi tro may be used to control diseases associated with cell immortality, such as neoplasia, and pathogenic parasites.

Applicant has determined that the inhibition of telomere shortening in a cell in vi tro is causally related to increasing the length of the replicative lifespan of that cell. Applicant has also determined that inhibition of telomerase activity in a cell in vi tro is causally related to reducing the ability of that cell to proliferate in an immortal manner. Thus, applicant is the first to provide data which clearly indicates that inhibition of telomere shortening in vivo or in vi tro, and that inhibition of telomerase activity in vivo or in vi tro, is therapeutically beneficial. Prior to applicant's experiments, as indicated above, there was no consensus by those in the art that one could predict that such experiments would provide the data observed by applicant, or that such manipulations would have therapeutic utility.

The invention also concerns the determination of cellular status by diagnostic techniques that analyze telomere length and telomerase activity, as a diagnostic of cellular capacity for proliferation. Assays for telomere length are performed to provide useful information on the relative age and remaining proliferative capability of a wide variety of cell types in numerous tissues. Sequences are also described from the telomeres of budding yeasts which are highly variable from strain to strain and provide sequences for oligonucleotide probes that would enable the rapid identification of yeast strains, and in the case of human and veterinary pathogens, the diagnosis of the strain of the pathogen. Telomerase activity and the presence of the enzyme is used as a marker for diagnosing and staging neoplasia and detecting pathogenic parasites. Applicant's experiments have, for the first time, determined a correlation between telomerase activity and the tumor cell phenotype, the hematopoetic stem cell phenotype, as well as a correlation between telomere length and the in vivo aged status of cells. As noted above, there was no consensus in the art that one could predict that such a relationship existed. In contrast, applicant has defined this relationship, and thus has now defined useful diagnostic tools by which to determine useful clinical data, such as to define a therapeutic protocol, or the futility of such a protocol to diagnose disease, or to predict the prognosis of a disease.

Thus, in a first aspect, the invention features methods for the treatment of a condition associated with cellular senescence or increased rate of proliferation of a cell (e.g. , telomere repeat loss associated with cell proliferation in the absence of telomerase) . A first method involves administering to the cell a therapeutically effective amount of an agent active to reduce loss of telomeric repeats during its proliferation. Such therapeutics may be especially applicable to conditions of increased rate of cell proliferatio .

By "increased rate of proliferation" of a cell is meant that the cell has a higher rate of cell division compared to normal cells of that cell type, or compared to normal cells within other individuals of that cell type. Examples of such cells include the CD4⁺ cells of HIV-infected individuals (see example below) , connective tissue fibroblasts associated with degenerative joint diseases, retinal pigmented epithelial cells associated with age-related macular degeneration, dermal fibroblasts from sun-exposed skin, astrocytes associated with Alzheimer's Disease and endothelial cells associated with atherosclerosis (see example below) . In each case, one particular type of cell or a group of cells is found to be replicating at an increased level compared to surrounding cells in those tissues, or compared to normal individuals, e.g. , in the case of CD4⁺ cells, individuals not infected with the HIV virus. Thus, the invention features administering to those cells an agent which reduces loss of telomere length in those cells while they proliferate, or reverses the loss by the re-expression of telomerase activity. The agent itself need not slow the proliferation process, but rather allow that proliferation process to continue for more cell divisions than would be observed in the absence of the agent. The agent may also be useful to slow telomere repeat loss occurring during normal aging (wherein the cells are proliferating at a normal rate and undergoing senescence late in life) , and for reducing telomere repeat loss while expanding cell number ex vivo for cell-based therapies, e.g. , bone marrow transplantation following gene therapy.

As described herein, useful agents can be readily identified by those of ordinary skill in the art using routine screening procedures. For example, a particular cell having a known telomere length is chosen and allowed to proliferate, and the length of telomere is measured during proliferation. Agents which are shown to reduce the loss of telomere length during such proliferation are useful in this invention. Particular examples of such agents are provided below. For example, oligonucleotides which are able to promote synthesis of DNA at the telomere ends are useful in this invention. In addition, telomerase may be added to a cell either by gene therapy techniques, or by introducing the enzyme itself or its equivalent into a cell, e.g. , by injection or lipofection. A second method for the treatment of cellular senescence involves the use of an agent to derepress telomerase in cells where the enzyme is normally repressed. Telomerase activity is not detectable in any normal human somatic cells other than certain hemapoietic stem cells in vi tro, but is detectable in cells that have abnormally reactivated the enzyme during the transformation of a normal cell into an immortal tumor cell. Telomerase activity may therefore be appropriate only in germ line cells and some stem cell populations such as hematopoetic stem cells. Since the loss of telomeric repeats leading to senescence in somatic cells is occuring due to the absence of adequate telomerase activity, agents that have the effect of activating telomerase would have the effect of adding arrays of telomeric repeats to telomeres, thereby imparting to mortal somatic cells increased replicative capacity, and imparting to senescent cells the ability to proliferate and appropriately exit the cell cycle (in the absence of growth factor stimulation with associated appropriate regulation of cell cycle-linked genes typically inappropriately expressed in senescence e.g., collagenase, urokinase, and other secreted proteases and protease inhibitors) . Such factors to derepress telomerase may be administered transiently or chronically to increase telomere length, and then removed, thereby allowing the somatic cells to again repress the expression of the enzyme utilizing the natural mechanisms of repression.

Such activators of telomerase may be found by screening techniques utilizing human cells that have the Ml mechanism of senescence abrogated by means of the expression of SV40 T-antigen. Such cells when grown to crisis, wherein the M2 mechanism is preventing their growth, will proliferate in response to agents that derepress telomerase. Such activity can be scored as the incorporation of radiolabeled nucleotides or proliferating clones can be selected for in a colony forming assay.

Such activators of telomerase would be useful as therapeutic agents to forestall and reverse cellular senescence, including but not limited to conditions associated with cellular senescence, e.g. , (a) cells with replicative capacity in the central nervous system, including astrocytes, endothelial cells, and fibroblasts which play a role in such age-related diseases as Alzheimer's disease, Parkinson's disease, Huntington's disease, and stroke, (b) cells with finite replicative capacity in the integument, including fibroblasts, sebaceous gland cells, melanocytes, keratinocytes, Langerhan's cells, and hair follicle cells which may play a role in age-related diseases of the integument such as dermal atrophy, elastolysis and skin wrinkling, sebaceous gland hyperplasia, senile lentigo, graying of hair and hair loss, chronic skin ulcers, and age-related impairment of wound healing, (c) cells with finite replicative capacity in the articular cartilage, such as chondrocyteε and lacunal and synovial fibroblasts which play a role in degenerative joint disease, (d) cells with finite replicative capacity in the bone, such as osteoblasts, bone marrow stromal fibroblasts, and osteoprogenitor cells which play a role in osteoporosis, (e) cells with finite replicative capacity in the immune system such as B and T lymphocytes, monocytes, neutrophils, eosinophils, basophils, NK cells and their respective progenitors, which may play a role in age-related immune system impairment, (f) cells with a finite replicative capacity in the vascular system including endothelial cells, smooth muscle cells, and adventitial fibroblasts which may play a role in age-related diseases of the vascular system including atherosclerosis, calcification, thrombosis, and aneurysms, and (g) cells with a finite replicative capacity in the eye such as pigmented epithelium and vascular endothelial cells which may play an important role in age-related macular degeneration.

In a second aspect, the invention features a method for treatment of a condition associated with an elevated level of telomerase activity within a cell. The method involves administering to that cell a therapeutically effective amount of an inhibitor of telomerase activity. The level of telomerase activity can be measured as described below, or by any other existing methods or equivalent methods. By "elevated level" of such activity is meant that the absolute level of telomerase activity in the particular cell is elevated compared to normal cells in that individual, or compared to normal cells in other individuals not suffering from the condition. Examples of such conditions include cancerous conditions, or conditions associated with the presence of cells which are not normally present in that individual, such as protozoan parasites or opportunistic pathogens, which require telomerase activity for their continued replication. Administration of an inhibitor can be achieved by any desired means well known to those of ordinary skill in the art.

In addition, the term "therapeutically effective amount" of an inhibitor is a well recognized phrase. The amount actually applied will be dependent upon the individual or animal to which treatment is to be applied, and will preferably be an optimized amount such that an inhibitory effect is achieved without significant side-effects (to the extent that those can be avoided by use of the inhibitor) . That is, if effective inhibition can be achieved with no side- effects with the inhibitor at a certain concentration, that concentration should be used as opposed to a higher concentration at which side-effects may become evident. If side-effects are unavoidable, however, the minimum amount of inhibitor that is necessary to achieve the inhibition desired may have to be used.

By "inhibitor" is simply meant any reagent, drug or chemical which is able to inhibit a telomerase activity in vi tro or in vivo . Such inhibitors can be readily identified using standard screening protocols in which a cellular extract or other preparation having telomerase activity is placed in contact with a potential inhibitor, and the level of telomerase activity measured in the presence or absence of the inhibitor, or in the presence of varying amounts of inhibitor. In this way, not only can useful inhibitors be identified, but the optimum level of such an inhibitor can be determined in vi tro for further testing in vivo.

One example of a suitable telomerase inhibitor assay is carried out in 96-well microtiter plates. One microtiter plate is used to make dilutions of the test compounds, while another plate is used for the actual assay. Duplicate reactions of each sample are performed. A mixture is made containing the appropriate amount of buffer, template oligonucleotide, and Tetrahymena or human telomerase extract for the number of the samples to be tested, and aliquots are placed in the assay plate. The test compounds are added individually and the plates are pre-incubated at 30°C. ³²P-dGTP is then added and the reaction allowed to proceed for 10 minutes at 30^*C. The total volume of each reaction is 10 μl . The reaction is then terminated by addition of Tris and EDTA, and half the volume (5 μl) spotted onto DE81 filter paper. The samples are allowed to air dry, and the filter paper is rinsed in 0.5 M NaPhosphate several times to wash away the unincorporated labeled nucleotide. After drying, the filter paper is exposed to a phosphor imaging plate and the amount of signal quantitated. By comparing the amount of signal for each of the test samples to control samples, the percent of inhibition can be determined.

Another example of a suitable telomerase inhibitor assay is carried out in 96-well microtiter plates. One microtiter plate is used to make dilutions of the test compounds, while another plate is used for the actual assay. Duplicate reactions of each sample are performed. A mixture is made containing the appropriate amount of buffer, nucleotides, biotintylated template oligonucleotide, and Tetrahymena or human telomerase extract for the number of the samples to be tested, and aliquots are placed in the assay plate. The test compounds are added individually. The reaction allowed to proceed for 60 minutes at 30°C. The total volume of each reaction is 40 μl . The reaction is then terminated, treated with proteinase K, transferred to a streptavadin coated microtiter plate and washed. Bound products are hybridized with 32-P labeled probe complementary to the extended telomeric sequences and washed extensively. Bound probe is then quantified and by comparing the amount of signal for each of the test samples to the control smaples, the percent of inhibition can be determined.

In addition, a large number of potentially useful inhibitors can be screened in a single test, since it is inhibition of telomerase activity that is desired. Thus, if a panel of 1,000 inhibitors is to be screened, all 1,000 inhibitors can potentially be placed into microtiter wells. If such an inhibitor is discovered, then the pool of 1,000 can be subdivided into 10 pools of 100 and the process repeated until an individual inhibitor is identified. As discussed herein, one particularly useful set of inhibitors includes oligonucleotides which are able to either bind with the RNA present in telomerase or able to prevent binding of that RNA to its DNA target or one of the telomerase protein components. Even more preferred are those oligonucleotides which cause inactivation or cleavage of the RNA present in a telomerase. That is, the oligonucleotide is chemically modified or has enzyme activity which causes such cleavage. The above screening may include screening of a pool of many different such oligonucleotide sequences. In addition, oligopeptides with random sequences can be screened to discover peptide inhibitors of telomerase or the orientation of functional groups that inhibit telomerase that, in turn, may lead to a small molecule inhibitor.

In addition, a large number of potentially useful compounds can be screened in extracts from natural products. Sources of such extracts can be from a large number of species of fungi, actinomyces, algae, insects, protozoa, plants, and bacteria. Those extracts showing inhibitory activity can then be analyzed to isolate the active molecule.

In related aspects, the invention features pharmaceutical compositions which include therapeutically effective amounts of the inhibitors or agents described above, in pharmaceutically acceptable buffers much as described below. These pharmaceutical compositions may include one or more of these inhibitors or agents, and be co-administered with other drugs. For example, AZT is commonly used for treatment of HIV, and may be co-administered with an inhibitor or agent of the present invention.

In a related aspect, the invention features a method for extending the ability of a cell to replicate. In this method, a replication-extending amount of an agent which is active to reduce loss of telomere length within the cell is provided during cell replication. As will be evident to those of ordinary skill in the art, this agent is similar to that useful for treatment of a condition associated with an increased rate of proliferation of a cell. However, this method is useful for the treatment of individuals not suffering from any particular condition, but in which one or more cell types are limiting in that patient, and whose life can be extended by extending the ability of those cells to continue replication. That is, the agent is added to delay the onset of cell senescence characterized by the inability of that cell to replicate further in an individual. One example of such a group of cells includes lymphocytes present in patients suffering from Downs Syndrome (although treatment of such cells may also be useful in individuals not identified as suffering from any particular condition or disease, but simply recognizing that one or more cells, or collections of cells are becoming limiting in the life span of that individual) .

It is notable that administration of such inhibitors or agents is not expected to be detrimental to any particular individual. However, should gene therapy be used to introduce a telomerase into any particular cell population, or other means be used to reversibly de-repress telomerase activity in somatic cells, care should be taken to ensure that the activity of that telomerase is carefully regulated, for example, by use of a promoter which can be regulated by the nutrition of the patient. Thus, for example, the promoter may only be activated when the patient eats a particular nutrient or pharmaceutical, and is otherwise inactive. In this way, should the cell population become malignant, that individual may readily inactivate telomerase of the cell and cause it to become mortal simply by no longer eating that nutrient or pharmaceutical.

In a further aspect, the invention features a method for diagnosis of a condition in a patient associated with an elevated level of telomerase activity within a cell. The method involves determining the presence or amount of telomerase within the cells in that patient.

In yet another aspect, the invention features a method for diagnosis of a condition associated with an increased rate of proliferation in that cell in an individual, or a condition in which the normal rate of proliferation has led to replicative senescence as a result of normal aging. Specifically, the method involves determining the length of telomeres within the cell.

Some of the various conditions for which diagnosis is possible are described above. As will be exemplified below, many methods exist for measuring the presence or amount of telomerase within a cell in a patient, and for determining the length of telomeres within the cell. It will be evident that the presence or amount of telomerase may be determined within an individual cell, and for any particular telomerase activity (whether it be caused by one particular enzyme or a plurality of enzymes) . Those in the art can readily formulate antibodies or their equivalent to distinguish between each type of telomerase present within a cell, or within an individual. In addition, the length of telomeres can be determined as an average length, or as a range of lengths much as described below. Each of these measurements will give precise information regarding the status of any particular individual.

Thus, applicant's invention has two prongs -- a therapeutic and a diagnostic prong. These will now be discussed in detail.

The therapeutic prong of the invention is related to the now clear observation that the ability of a cell to remain immortal lies in the ability of that cell to maintain or increase the telomere repeat length of chromosomes within that cell. Such a telomere repeat length can be maintained by the presence of sufficient activity of telomerase, or an equivalent enzyme, within the cell. Thus, therapeutic approaches to reducing the potential of a cell to remain immortal focus on the inhibition of telomerase or equivalent activity within those cells in which it is desirable to cause cell death. Examples of such cells include cancerous cells, which are one example of somatic cells which have regained the ability to express telomerase, and have become immortal. Applicant has now shown that such cells can be made mortal once more by inhibition of telomerase activity. As such, inhibition can be achieved in a multitude of ways including, as illustrated below, the use of oligonucleotides which, in some manner, block the ability of telomerase to extend telomeres in vivo .

Thus, oligonucleotides can be designed either to bind to a telomere (to block the ability of telomerase to bind to that telomere, and thereby extend that telomere) , or to bind to the resident oligonucleotide (RNA) present in telomerase to thereby block telomerase activity on any nucleic acid

(telomere) or to the mRNA encoding telomerase protein components to block expression of those proteins and hence telomerase activity. Such oligonucleotides may be formed from naturally occurring nucleotides, or may include modified nucleotides to either increase the stability of the therapeutic agent, or cause permanent inactivation of the telomerase, e.g.. the positioning of a chain terminating nucleotide at the 3' end of the molecule of a nucleotide with a reactive group capable of forming a covalent bond with telomerase. Such molecules may also include ribozyme sequences. In addition, non-oligonucleotide based therapies can be readily devised by screening for those molecules which have an ability to inhibit telomerase activity in vitro, and then using those molecules in vivo . Such a screen is readily performed and will provide a large number of useful therapeutic molecules. These molecules may be used for treatment of cancers, of any type, including solid tumors and leukemias (including those in which cells are immortalized, including: apudoma, choristoma, branchioma, malignant carcinoid syndrome, carcinoid heart disease, carcinoma (e.g.. Walker, basal cell, basosquamous, Brown-Pearce, ductal, Ehrlich tumor, in situ, Krebs 2, merkel cell, mucinous, non-small cell lung, oat cell, papillary, scirrhous, bronchiolar, bronchogenic, squamous cell, and transitional cell) , histiocytic disorders, leukemia

(e.g. , b-cell, mixed-cell, null-cell, T-cell, T-cell chronic, HTLV-II-associated, lyphocytic acute, lymphocytic chronic, mast-cell, and myeloid) , histiocytosis malignant, Hodgkin's disease, immunoproliferative small, non-Hodgkin's lymphoma, plasmacytoma, reticuloendotheliosis, melanoma, chondro lastoma, chondroma, chondrosarcoma, fibroma, fibrosarcoma, giant cell tumors, histiocytoma, lipoma, liposarcoma, mesothelioma, myxoma, myxosarcoma, osteoma, osteosarcoma, Ewing's sarcoma, synovioma, adenofibroma, adenolymphoma, carcinosarcoma, chordoma, craniopharyngioma, dysgerminoma, hamartoma, mesenchymoma, mesonephroma, myosarcoma, ameloblaεtoma, cemento a, odontoma, teratoma, thymoma, trophoblastic tumor, adenocarcinoma, adenoma, cholangioma, cholesteatoma, cylindroma, cystadenocarcinoma, cystadenoma, granulosa cell tumor, gynandroblaεtoma, hepatoma, hidradenoma, islet cell tumor, leydig cell tumor, papilloma, sertoli cell tumor, theca cell tumor, leiomyoma, leiomyosarcoma, myoblastoma, myoma, myoεarcoma, rhabdomyoma, rhabdomyosarcoma, ependymoma, ganglioneuroma, glioma, medulloblastoma, meningioma, neurilemmoma, neuroblaεtoma, neuroepithelioma, neurofibroma, neuroma, paraganglioma, paraganglioma nonchromaffin, angiokeratoma, angiolymphoid hyperplaεia with eosinophilia, angioma scleroεing, angiomatoεiε, glomangioma, hemangioendothelioma, hemangioma, hemangiopericytoma, hemangiosarcoma, lymphangioma, lymphangiomyoma, lymphangiosarcoma, pinealoma, carcinoεarcoma, chondroεarcoma, cyεtoεarcoma phyllodeε, fibroεarcoma, hemangioεarcoma, leiomyoεarcoma, leukosarcoma, liposarcoma, lymphangiosarcoma, myosarcoma, myxosarcoma, ovarian carcinoma, rhabdomyosarcoma, sarcoma (e.g. , Ewing'ε, experimental, Kapoεi'ε, and mast-cell), neoplasms (e.g. , bone, breast, digeεtive εyεtem, colorectal, liver, pancreatic, pituitary, testicular, orbital, head and neck, central nervous system, acoustic, pelvic, reεpiratory tract, and urogenital) , neurofibromatoεis, and cervical dysplasia) , and for treatment of other conditions in which cells have become immortalized.

Applicant has also determined that it iε important to εlow the loεε of telomere εequences, in particular, cells in aεεociation with certain diseases (although such treatment iε not limited to this, and can be used in normal aging and ex vivo treatments) . For example, some diseaseε are manifest by the abnormally faεt rate of proliferation of one or more particular groupε of cellε. Applicant haε determined that it is the senescence of those groupε of cellε at an abnormally early age that eventually leadε to diεeaεe in that patient. One example of εuch a disease is AIDS, in which death is caused by the early senescence of CD4⁺ cells. It is important to note that such cells age, not because of abnormal amount of loεε of telomere εequenceε per cell doubling (although thiε may be a factor) , but rather becauεe the replicative rate of the CD4⁺ cellε is increased εuch that telomere attrition iε caused at a greater rate than normal for that group of cells. Thus, applicant provides therapeutic agentε which can be uεed for treatment of such diseases, and also provides a related diagnostic procedure by which similar diseases can be detected so that appropriate therapeutic protocols can be devised and followed.

Specifically, the losε of telomereε within any particular cell population can be reduced by proviεion of an oligonucleotide which reduces the extent of telomere attrition during cell division, and thus increaseε the number of cell divisions that may occur before a cell becomes senescent. Other reagents, for example, telomerase, or its mRNAs or its geneε, may be provided to a cell in order to reduce telomere loεs, add telomeric repeats, or to make that cell immortal. Other enzymatic activities may be used to enhance the lengthening of telomeres within εuch cells, for example, by providing certain viral reverse tranεcriptaεeε and an RNA template for the C-rich telomerase repeat sequence which can function to syntheεize telomere εequences within a cell. In addition, equivalent εuch moleculeε, or other molecules may be readily screened to determine those that will reduce loεε of telomereε or activate telomeraεe. Such εcreenε may occur in vi tro, and the therapeutic agents discovered by such screening utilized in the above method in vivo . Other therapeutic treatments relate to the finding of unuεual telomeric DNA εequenceε in a group of fungi, εpecifically a group of budding yeaεtε that includeε some pathogens - Candida albicans, Candida tropicalis and Candida paratropicaliε - aε well aε nonpathogenic fungi. Theεe reεultε are deεcribed in more detail below. Drugs or chemical agents can be used to specifically exploit the unusual nature of the telomeric DNA of fungi. This includes the introduction of antiεense polynucleotides εpecific to the telomeric repeat DNA εequences, in order to block telomere synthesis in these and any related pathogenε. Such a block will lead to fungal death.

Thiε approach iε advantageous becauεe of the unuεual nature of the telomeric DNA in theεe fungi. The unuεually high DNA εequence complexity of the telomeric repeatε of these fungi provides specificity, and potential for minimal side effects, of the antifungal agent or the antisenεe DNA or RNA.

Agents that are potentially useful antifungal agents include: AZT, d4T, ddl, ddC, and ddA. The telomere εynthesis of theεe fungi iε expected to show differential inhibition to these drugs, and in some caseε to be more sensitive than the telomere syntheεiε in the human or other animal or plant host cells.

We performed a preliminary test of the use of antisense techniques in living fungal cellε. A εtretch of 40 bp of telomeric DNA εequence, imbedded in a conεerved εequence flanking a region of Candida albicans chromosomal DNA, waε introduced on a circular molecule into Candida albicanε cells. The transformed cells had high copy numbers of the introduced telomeric DNA sequence. 10% of the transformants exhibited greatly (~ 3 - fold) increased length of telomeric DNA. This result indicates that telomeric DNA can be modulated in vivo by introduction of telomeric sequence polynucleotides into cells. This demonstrates the need to test a particular oligonucleotide to ensure that it haε the deεired activity.

With regard to diagnostic procedures, examples of εuch procedureε become evident from the discusεion above with regard to therapy. Applicant haε determined that the length of the telomere iε indicative of the life expectancy of a cell containing that telomere, and of an individual composed of εuch cells. Thus, the length of a telomere is directly correlated to the life span of an individual cell. As discusεed above, certain populationε of cellε may loεe telomereε at a greater rate than the other cells within an individual, and those cells may thuε become age-limiting within an individual organism. However, diagnostic procedures can now be developed (as described herein) which can be used to indicate the potential life span of any individual cell type, and to follow telomere loεs so that a revised estimate to that life span can be made with time.

In certain diseaεeε, for example AIDS, aε diεcuεεed above, it would, of course, be important to follow the telomere length in CD4⁺ cells and cells εharing itε hematopoietic lineage. In addition, the recognition that CD4⁺ cellε are limiting in such individuals allows a therapeutic protocol to be devised in which CD4⁺ cells can be removed from the individual at an early age when AIDS is first detected, stored in a bank, and then reintroduced into the individual at a later age when that individual no longer has the required CD4⁺ cellε available. Theεe cellε can be expanded in number in the preεence of agentε which εlow telomere repeat loss, e.g. , C-rich telomeric oligonucleotides or agents to transiently de-repress telomerase to ensure that cells re-administered to the individual have maximum replicative capacity. Thus, an individual's life can be extended by a protocol involving continued administration of that individual's limiting cells at appropriate time points. Theεe appropriate pointε can be determined by following CD4⁺ cell senescence, or by determining the length of telomeres within such CD4⁺ cellε (as an indication of when those cells will become senescent) . In the case of AIDS, there may be waves of senescent telomere length in peripheral blood lymphocytes with bone marrow stem cells still having replicative capacity. In thiε way, rather than wait until a cell becomeε εeneεcent (and thereby putting an individual at riεk of death) telomere length may be followed until the length iε reduced below that determined to be pre-senescent, and thereby the timing of administration of new CD4⁺ cells or colony stimulating factors can be optimized.

A number of similar therapeutic protocols can be used. Early passage cells (i.e., cells which have undergone few divisions, and thus have long telomereε) can be iεolated from the tiεεue of donors, and prepared for reintroduction to the donor. The cells with the greatest replicative capacity can be isolated by using telomere length as a marker of replicative capacity. The cells can then be grown up in a culture medium which slows the replicative seneεcence of theεe cells.

For example, such a medium could contain a C-rich (CTR) terminal repeat sequence. This oligonucleotide slows the losε of telomere repeats and extends the replicative capacity of cells. Such growth is beneficial becauεe in the abεence of factorε which εlow cellular senescence, the cellε would εenesce in vi tro . In addition, telomerase activity can be added to εuch cellε to increase telomerase length and thereby increase the replicative capacity of the cells. Thiε procedure can be applied to εeveral different tiεεueε. For example, this therapeutic procedure could be applied to bone marrow stem cells, which applicant believes have finite replicative capacity. Numerous kinds of ex-vivo cell therapies using bone marrow stem cellε are currently under development. Many of theεe are deεigned in order to perform gene therapy on the explanted cellε, expand the clones that have incorporated the genetic construct, and then to reintroduce the altered cells. The procedure described above allows one to iεolate the stem cells with the introduced construct which have the greatest replicative capacity, and thus would reduce the conεequenceε of replicative senescence. Since bone marrow stem cellε and related hematopoietic εtem cells posεeεs telomerase activity (Fig. 41) telomerase activity provideε a novel meanε of identifying theεe stem cells in a mixed population of bone marrow or peripheral blood cells.

This procedure as applied to bone marrow stem cells is also of benefit apart from gene therapy protocols. For example, in caseε where an individual is suffering from a disease linked to an immune system undergoing replicative senescence, e.g. normal aging, or cases where the immune system has been severely and chronically stresεed, e.g. HIV infection, it may be deεirable to iεolate bone marrow stem cells, amplify them in the presence of factors that εlow or reverse replicative seneεcence, and reintroduce them to reconεtitute the immune εyεtem. Other examples include treatment of muscular dystrophy by use of muscle εatellite cellε treated aε described herein. The described therapeutic procedure for the preparation of cellε for reintroduction to donors can also be applied to dermal fibroblasts. Young or early pasεage fibroblaεtε can be isolated from old by means of monoclonal antibodies or electrophoretic mobility and a computerized laser scanner (e.g., ACAS Machine 570 Interactive Laser Cytometer manufactured by Meridian Inεtrumentε, Inc.). The replicative capacity of clones of theεe cells can then be determined by either of two methods. The first of theεe methods uses telomere length to predict replicative capacity, as deεcribed above. In the εecond method, the iεolated fibroblaεtε are aεεayed for relative levelε of collagenase activity or other gene productε altered with cell εenescence (e.g., stromelysin, plasminogen activator, lysosomal hydrolaseε εuch aε β-Ω- galadoεidaεe, EPC-1) . Cellular senescence of dermal fibroblastε correlates with an increased production of collagenase activity. Thus, the clones of cells with the greateεt replicative capacity can be identified by either of theεe methods. The cells can then be subcultured in a culture medium which εlowε the replicative senescence of these cellε until sufficient numbers of cells are obtained. The cells are then recombined with autologous matrix proteins obtained from theεe cellε, and the reεulting living cell/protein matrix is injected into dermal skin wrinkles for the permanent restoration of skin contour. This method has the advantage of removing the poεεibility of immune rejection of foreign protein or heterologouε cellε. Alεo, the incluεion of selected young cells will stabilize the injected matrix in a manner similar to the way young cellε normally maintain dermal protein in young εkin. Such young cellε have low proteinase activity and thus are less likely to destroy the matrix needed to maintain the cell structure. This procedure can also be applied to the preparation of young skin matrix to be implanted in regions of burned skin to improve wound healing.

This procedure can also be used to isolate early passage cellε for cell-baεed therapieε from other tissues, for example, osteoblastε to treat oεteoporoεis, retinal pigmented epithelial cells for age-related macular-degeneration, chondroctes for osteoarthritiε, and εo on.

Thuε, the diagnostic procedures of this invention include procedureε in which telomere length in different cell populations is measured to determine whether any particular cell population is limiting in the life εpan of an individual, and then determining a therapeutic protocol to insure that such cells are no longer limiting to that individual. In addition, such cell populations may be εpecifically targeted by εpecific drug administration to insure that telomere length loss is reduced, as discussed above.

Other diagnostic procedures include meaεurement of telomerase activity as an indication of the presence of immortal cells within an individual. A more precise measurement of such immortality iε the presence of the telomerase enzyme itself. Such an enzyme can be readily detected using standard procedureε, including assay of telomerase activitieε, but alεo by use of antibodies to telomerase, or by use of oligonucleotideε that hybridize to the nucleic acid (template RNA) present in telomerase, or DNA or RJSTA probes for the mRNAs of telomerase proteins. Immunohistochemical and in εitu hybridization techniqueε allow the preciεe identification of telomeraεe poεitive cells in histological specimens for diagnostic and prognostic tests. The presence of telomerase is indicative of cells which are immortal and frequently metastatic, and such a diagnostic allows pinpointing of such metastatic cells, much as CD44 iε alleged to do. See, Leff, 3(217) BioWorld Today 1, 3, 1992.

It is evident that the diagnostic procedures of the present invention provide the first real method for determining how far certain individuals have progressed in a certain disease. For example, in the AIDS disease, this is the first effective methodology which allows prior determination of the time at which an HIV positive individual will become immunocompromised. Thiε information is useful for determining the timing of administration of prophylaxis for opportunistic infections such aε ketoconazole adminiεtration, and will aid in development of new drug regimenε or therapieε. In addition, the determination of the optimum timing of adminiεtration of certain drugε will reduce the coεt of treating an individual, reduce the opportunity for the drug becoming toxic to the individual, and reduce the potential for the individual developing reεiεtance to such a drug.

In other related aspects, the invention features a method for treatment of a disease or condition associated with cell senescence, by administering a therapeutically effective amount of an agent active to derepress telomerase in seneεcing cell . A related aεpect involves screening for a telomerase derepresεion agent by contacting a potential agent with a cell lacking telomerase activity, and determining whether the agent increases the level of telomerase activity, e.g. , by using a cell expreεsing an inducible T antigen. Such an assay allows rapid screening of agents which are present in combinatorial librarieε, or known to be carcinogens.

Applicant recognizes that known agents may be useful in treatment of cancers since they are active at telomerase itεelf, or at the gene expressing the telomerase. Thuε, εuch agentε can be identified in this invention aε uεeful in the treatment of diεeases or conditions for which they were not previously known to be efficaciouε. Indeed, agentε which were previously thought to lack utility because they have little if any effect on cell viability after only 24-48 hours of treatment, can be shown to have utility if they are active on telomeraεe in vivo, and thuε affect cell viability only after several cell divisions.

Other features and advantages of the invention will be apparent from the following description of the preferred embodiments thereof, and from the claims. Description of the Preferred Embodiments The drawings will first briefly be described.

Drawings

Figε. 1-3 are graphε where the cell type and/or the culture conditionε are varied, plotting dayε in culture (horizontal axiε) length verεuε cell number (vertical axis) .

Fig. 4 is a linear plot of mean terminal reεtriction fragment (TRF) length versus PDL for human umbilical vein endothelial cell cultures . The plot had a slope (m) of -190 ± 10 bp/PD, r=-0.98, P=0.01.

Fig. 5 is a plot of mean TRF of endothelial cell cultures from human iliac arteries and iliac veins as a function of donor age. Parameters for iliac arteries are: m=-102 bp/yr, r=-0.98, P=0.01 and for iliac veins are: m=-42 bp/yr, r=-0.71, P=0.14.

Fig. 6 is a plot of decrease in mean TRF of medial tissue from the aortic arch, abdominal aorta, iliac artery and iliac vein as a function of donor age. Parameters for linear plot are: m=-47 bp/yr, r=-0.85, P=0.05.

Fig. 7 is a plot of mean TRF length from PBLε plotted aε a function of donor age. The εlope of the linear regression line (-41 ± 2.6 bp/y) is significantly different from 0 (p<0.00005) .

Fig. 8 is a plot showing accelerated telomere losε in Down's Syndrome (DS) patients. Genomic DNA isolated from PBLs of DS patientε waε analyzed aε deεcribed in Fig. 7. Mean TRF length is shown as a function of donor age, for DS patientε (open squares) , and age-matched controls (solid squares) . The slope of the linear regresεion lineε (-133 ± 15 bp/y, trisomy, vs -43 ± 7.7, normalε) are εignificantly different (p<0.0005) .

Fig. 9 iε a plot εhowing decrease in mean TRF length in cultured T-lymphocytes as a function of population doubling (shown for DNA from two normal individuals) . Donor ages for these cells were not available. The slopeε of theεe lineε (-80 + 19 (o) and -102 ± 5.4 (0) bp/doubling) are εignificantly different from zero (p<0.0001) . Mean TRF length at terminal passage from a third donor for which multiple passageε were not available iε also shown (upsidedown V-symbol) .

Fig. 10 is a copy of an autoradiogram showing

TRF lengths of ovarian carcinoma and control normal cells. DNA from cells in ascitic fluid from 2 patientε

(cas and wad) was digested with Hinfl and Rsal separated by electrophoresis, hybridized to the telomeric probe ³²P (CCCTAA)₃, stringently washed and autoradiographed. The cells of ascitic fluid from 7 other patients were separated into adhering normal cells (N) and tumour clumps in the media (T) . The DNA was extracted and run as above. DNA from patient was obtained from both the first and forth paracenteεis . Tumour cells from patients were cultured and DNA was obtained at the respected population doubling (pd) .

Fig. 11 showε telomeraεe activity in ovarian carcinoma cells. S100 extracts from the previously studied transformant cell line 293 CSH, the tumor cell line HEY, purified tumour cell population and cells directly from the ascitic fluid from patients were incubated with the telomere primer (TTAGGG)₃ in the presence of dATP and TTP, ³²PdGTP and buffer. The reaction products were separated on a sequencing gel and exposed to a Phospholmager screen. Either single (1) or double reactions (2) were tested.

Fig. 12 is a copy of an autoradiogram showing TRF lengths in HME-31 cellε and HME31-E6 cellε to extended lifeεpan (PD68) and subεequent immortalization and εtabilization of telomere length (PD81, 107) . Fig. 13 is a copy of an autoradiogram showing the effect of CTO on telomere length during the senescence of HME31:E6 cells. An intermediate time point is chosen to show the dose-dependent protective effect of CTO oligonucleotide. Fig. 14 iε a graph εhowing extenεion of the life εpan of IMR90 lung fibroblaεt cellε in reεponse to the CTO oligonucleotide. Figs. 15 and 16 are copies of autoradiograms showing the effect of GTO on telomere length in IDH4 cells.

Fig. 17 is a graph showing extension of the life span of HME31:E6 human breast epithelial cells in responεe to the^ CTO oligonucleotide.

Fig. 18A. εhowε the templating portion of the

Tetrahymena telomeraεe RNA with reεidues numbered 1

(5') through 9 (3') below it. The oligonucleotide primer with the sequence T₂G₄T₂G₄ binds to the template by the base-pairing shown. Elongation followed by template translocation are thought to occur aε indicated.

Fig. 18B εhowε positions of major chain termination on the telomerase RNA template by different nucleoεide triphoεphate analog . The telomeraεe RNA template sequence is εhown as in Fig. 18A. Arrows indicate the poεition of maximal chain termination for each nucleoεide triphoεphate (derived from the nucleoεide) analog shown.

Fig. 19A-F are graphs showing that nucleoside analog triphoεphateε inhibit incorporation of a ³²P label in a Tetrahymena telomerase assay. The effect of adding increasing concentrations of the analog, unlabeled dGTP or unlabeled TTP on the incorporation of labeled nucleotides was measured uεing a quantitative telomerase reaction assay. Radioactivity incorporated (cpm) was plotted against the concentration of competitors indicated in each panel. (A. labeled with [α--³²P]TTP. B-F. labeled with [α^*-³²P]dGTP. F. Effect of streptomycin sulfate on the telomerase reaction. The incorporation in the presence of 40 mM sodium sulfate iε εhown aε the control for εtreptomycin εulfate) .

Fig. 20A and B show the effect of nucleoside triphosphate analogε on pauεing patterns and procesεivity of telomeraεe in vi tro . Specifically, Fig. 20A εhowε telomeraεe reactionε in the preεence or absence of the indicated nucleoside triphoεphate analog . Unlabeled TTP competitor waε also analyzed as a control, with and without primer in the reaction mix. Productε were then analyzed on a denaturing polyacrylamide gel. Fig 2OB εhowε εtandard telomeraεe reactionε were performed in the preεence of ddGTP (lanes 4-6) , ddlTP (lanes 7-9), or DMSO (lane 1) . DMSO was the solvent for ddGTP and at the highest concentration tested (1%) εhowed no effect on the reactionε compared with control reactionε run without analog or DMSO (control laneε 2-3) . Productε were analyzed on a denaturing polyacrylamide gel .

Fig. 21A-D εhowε Southern blot analysis to demonεtrate the effect of nucleoεide analogε on telomere length in vivo. using a nick-translated [cx- ³ P] -labeled plasmid containing a 3' rDNA fragment as probe. Genomic DNA was digeεted with Pεtl and BamHI and the rDNA telomeres analyzed. The telomeric Pstl fragment from the rDNA is between the 1.6 and 1.0 kb markers, indicated as lineε on both sides of each panel. The constant 2.8 kb band is the adjacent internal Pstl rDNA fragment. Specifically, Fig. 21A showε reεultε with a clone of Tetrahymena thermophila grown in 2% PPYS in the abεence (-) and three cloneε in the presence (+) of 5 mM AZT. Each set of three lanes shows the results for a single cell clone grown vegetatively and transferred after 3 days (lanes 1, 4, 7, 10), 10 dayε (laneε 2, 5, 8, 11) and 16 dayε (laneε 3,6,9,12) . Fig. 21B εhowε that growth in different concentrations of AZT consiεtently reεulted in concentration-dependent εhortening of telomereε in log phaεe cellε grown in thymine-deficient broth (Iεobroth) pluε AZT. DNA made from cellε εampled at 6, 10, and 16 dayε εhow that εhortened telomere lengthε remain conεtant between 6 and 16 dayε in culture. Laneε 1, 5, 9, 0 mM AZT control; lanes 2, 6, 10, 0.01 mM AZT; laneε 3, 7, 11, 0.lmM AZT; lanes 4, 8, 12, 1 mM AZT. Fig. 21C shows cells grown vegetatively in 2% PPYS with no addition (lane 1) , with 1% DMSO, the solvent for Ara-G, ("C", lanes 2 and 5), and with Ara-G (lane 3, lmM ; lanes 4 and 6, 2mM ) at 14 and 27 days in culture. Fig 21D shows analysiε of DNA from εingle- cell cultureε grown in Isobroth plus 1 mM AZT (lanes 2 and 3) segregated into two classes based on growth rate: "slow" ( "S", 0-1 doubling per day, lane 2) or "faεt" ( "F", 2-4 doubling per day, lane 3) . DNA from control cultures grown in the absence of AZT are indicated ("C", 2-4 doubling per day, lane 1) . Several cultures were pooled in order to obtain sufficient DNA for analyεis.

Fig. 22 showε PCR analyεis of DNA from Tetrahymena cellε conjugated in the preεence of analog and starved for the duration of mating. A Telomeric primer and a 5' rDNA primer were used in PCR reactions with DNA from cells conjugated in the presence or abεence of analog to detect the addition of telomereε to the 11Kb rDNA formed during macronuclear development. A reaction waε run without DNA as a control. Tests included use of 5 mM AZT; 1 mM Ara-G, and 1 mM Acyclo-G. SB210 cellε were alεo mock- conjugated as a control. The expected product is approximately 1400 bp. In addition, 3' micronuclear rDNA primers were used on the same DNA to demonstrate the presence and competence of the DNA samples for PCR. The expected band is 810 bp. In the figure southern blot analyεiε of the 5' rDNA telomeric PCR reactions using a random-primed ³²P-labeled 5' rDNA probe confirmed the 1400 bp PCR product as part of the 5'rDNA with telomeres, from the 11Kb rDNA specieε formed transiently during macronuclear development. No hybridization is seen in the no DNA control (lane 1) or the SB210 mock-conjugated control (lane 6) . Lane 2, no added analog; lane 3, 5 mM AZT., lane 4, 1 mM Ara-G; lane 5, 1 mM Acyclo-G; lane 6, mock-conjugated SB210 cell DNA. These resultε were reproduced in three separate experiments.

Fig. 23 shows growth of cultured JY lymphoma cells with RPMI medium and no added agents (control) and with a relatively low doεe of ddG, AZT, ara-G, and ddl. The DMSO iε a control for ddG.

Fig. 24 shows the growth of cultured JY lymphoma cells cultured in an analogous manner to those in Fig. 23, but treated with relatively higher doses of potential telomerase inhibitors.

Fig. 25 showε Southern blot of DNA iεolated from JY lymphoma cellε at weeks one and three probed with the telomeric repeat sequence (TTAGGG)₃. The first lane iε DNA from the cellε at the εtart of the experiment, the second is the RPMI control, and the third is cellε treated with AZT for the times indicated.

Fig. 26 shows fibroblast DNA hybridized by Southern blot to the telomeric (TTAGGG)₃ probe. Lane labeled "Hinfl" iε DNA digeεted with the reεtriction enzyme Hinfl, the lane labeled "O" had no treatment, the lane labeled "P only" waε treated with piperidine, and the lane labelled "P + DMS" waε piperidine and dimethyl sulfate treated. Fig. 27 shows the inhibition of human telomerase achieved by the agent ddG at various doεageε in three εeparate experimentε. The telomeraεe was derived from the tumor cell line 293.

Fig. 28 shows hybridization of C. albicans telomeric repeats to genomic DNAs of a variety of other Candida specieε. Genomic DNAε of eight εpecies of yeasts were digeεted with EcoRl, electrophoresed on 0.8% agarose, blotted, and then probed with a ³²P- labeled telomeric fragment from C. albicans WO-1. Hybridization was carried out at 55° C and washeε were at the εame temperature in Na₂HP0₄ at 200mM Na⁺ and 2% SDS. DNA size markers, meaεured in kilobase pairs (kb) , are shown at the right. The specieε used here are C. guillermondii , S. cerevisiae, C. pseudotropicalis, Kluyveromyces lactis, C. lusi taniae, C. mal tosa, C. tropicalis, and C. albicans . Asteriεks indicate particular strainε from which telomereε were cloned. Strainε beginning with "B" are N.I.H. εtrains obtained from B. Wickes.

Fig. 29 εhowε Bal31 εenεitivity of genomic copieε of the tandem repeatε in K. lactis ATCC 32143 (left panel) and C. guillermondii B-3163 (right panel) . Uncut yeast genomic DNAs were incubated with Bal31 nuclease for increasing periods of time (given in minutes above each lane) , then digested with EcoRl and electrophoresed on a 0.8% agaroεe gel, and blotted onto a nylon membrane. For K. lactis, probing was done with a ³²P-kinased 25 baεe oligonucleotide identical in εequence to the K. lactis telomeric repeat shown in Fig. 30. Hybridization and washes were carried out at 49°C. For C. guillermondii , probing was done with ³²P- labeled pCgui3, a pBluescript vector (Stratagene, LaJolla, CA) carrying a α-2 kb telomeric clone from C. guillermondii . Hybridization and washing (in 200 mM Na⁺) were carried out at 54°C. Most bands are gone by the 1 min. time point. Approximately three other bands are shortening but are not gone at 3 min. These latter bands presumably are homologous to the particular εubtelomeric sequences present in pCgui3. DNA size markers (in kb) are indicated at the right of each panel. Fig. 30 shows sequenceε of telomeric repeatε from εeveral budding yeaεt εpecies. Specifically, telomere-enriched libraries were constructed from genomic DNA by standard methods . Uncut yeast genomic DNA was ligated to a blunt-ended linearized plasmid vector and then thiε ligated mix was digested with a restriction enzyme that cleaveε both within the vector' ε polylinker and within a few kilobases of at least some of the putative telomeric endε of the εpecieε in question. No enzymatic pre-treatment was done to produce blunt-ends of the telomeres in the genomic DNA prior to the initial ligations . Plasmidε were then recircularized with T4DNA ligase, and transformed into E. coli cellε prior to εcreening for putative telomere cloneε by colony hybridization. The librarieε from C. maltoεa. C. pεeudotropicaliε, two εtrainε of C. tropicaliε, and K. lactis ATCC 32143, species which εhowed multiple bandε that croεε hybridized to the C. albicanε telomeric repeat probe, were εcreened with thiε probe. A cloned S. cereviεiae telomere probe (repeat unit TG₂_₃ (GT) ₁._3ι) waε used to screen the telomere - enriched library from C _ glabrata, whose genomic DNA cross - hybridized with this, but not with the C. albicans telomeric repeat probe. C. guillermondii DNA did not appreciably cross- hybridize with either the C. albicans or the S . cerevisiae telomeric probeε at the stringencies tested. The telomere - enriched library from this species was screened using total genomic C. guillermondii DNA aε a probe. Thiε procedure can be uεed to identify all cloneε containing repetitive sequences and we reasoned that telomeres should be a reasonable percentage of the repetitive sequences found in telomere enriched libraries. Typically, a few hundred E. coli transformants were obtained for each small library and up to nine putative telomere clones were obtained from each. Nine repetitive DNA clones were obtained from C___. guillermondii, three of which proved to be telomeric.

Fig. 31 shows two types of telomeric repeats preεent in certain C. tropicaliε εtrainε . Genomic DNAε from ten (only five here are εhown) C. tropicaliε εtrainε and C. albicans WO - 1 were digested with Clal. eletrophoresed on a 0.8% agarose gel, blotted, and probed with oligonucleotides specific to either the "AC form of C. tropicaliε telomeric repeat (left panel) or to the "AA" form of repeat (right panel) . Sequences of these two oligonucleotideε are: 5'ACGGATGTCACG ("AC") and 5'GTGTAAGGATG ("AA") with the poεition of the dimorphic baεe shown underlined. Hybridization with the kinased "AC" probe was at 47^*C, and hybridization with the "AA" probe at 24^* C. Washes for both were in 2% SDS with 500 mM Na⁺. The specificity of the "AA" probe iε indicated by its failure to hybridize with the C. albicans telomeres, despite only one base mismatch and the fact that the C. albicans cells used here have much longer telomeres (and therefore many more telomeric repeats) than do C. tropicalis strains. The shortness of the C. tropicalis telomeres may explain why they appear to be particularly homogeneous in εize, as iε suggested by the relative sharpness of individual telomeric bands .

Fig. 32 shows a Southern blot of DNA isolated from JY cells hybridized to the (TTAGGG)₃ probe. Cells were treated over a 10 week period with either lOμM ddG in 0.01% DMSO or medium with 0.01% DMSO only. Cells treated with ddG showed a marked decrease in mean telomere length consiεtant with the inhibition of telomeraεe activity.

Fig. 33 shows telomerase activity in cells from aεcitic fluid. Specifically, S100 extracts were prepared, protein concentrations determined and telomerase activity asεayed by incubating S100 extractε with an equal volume of reaction mixture containing buffer, telomere primer (TTAGGG)₃, c-³²PdGTP, TTP and dATP, at 30°C for 1 hour. The reactionε were terminated with RNaεe followed by deproteination with proteinaεe K. Unincorporated α.³²PdGTP was removed using NICK SPIN columns (Pharmacia) according to the supplier's direction. Products were resolved on a sequencing gel and expoεed to either a Phoεphorlmager screen (Molecular Dynamics) . A ladder (L) and kinased 5'³ P(TTAGGG)₃ (0) were run as markers. Fig. 33A shows telomeraεe aεεayed in SIOO extractε with equal protein concentration (<=11 mg/ml) prepared from the control human cell line 293 CSH, a εubline of 293 cell line, and from unfractionated aεcitic fluid cellε from patient Dem-1 and Rud-1. In laneε 1, 3 and 5 RNaεe waε added to the extractε prior to addition of α.³²PdGTP. Fig. 33B showε SIOO extractε iεolated and aεsayed for telomerase activity from the early passage cultures of cells from patients Pres-3 and Nag-1 compared to 293 cellε. All extractε were aεεayed at a protein concentration of « 2-3 mg/ml.

Fig. 34 iε a diagrammatic representation of oligonucleotides useful in a PCR asεay for telomerase activity, and their primer extension products. Specifically, sequences of two telomeric oligonucleotide substrateε [(TTAGGG)₃, (GTTAGG)₃] and non-telomeric oligonucleotide εubεtrate/primer (M2) are εhown. Predicted telomerase products for each oligo substrateε are alεo shown with vertical broken lines dividing each telomeric repeats. Upstream/substrate

(M2) and downstream (CX) primers used for the PCR-based aεεay are εhown with the direction of poly eraεe reaction (arrowε) and the potential binding εiteε for the downstream CX primer (solid vertical lines) . Fig. 35 is a copy of an autoradiogram showing the

PCR products using various subεtrate oligonucleotides. Specifically, conventional telomerase aεεays were performed on the two telomeric oligo substrate and M2 substrate primer, run on 8% polyacrylamide sequencing gel, and exposed overnight on a phosphori ager. Lane

M is a synthetic marker corresponding to the first, second, and third telomerase productε from (TTAGGG)₃ oligo substrates.

Fig. 36 is a copy of autoradiograms showing asεay reεults with a conventional assay compared to a PCR assay. Specifically, telomerase products (1/10) from the conventional telomeraεe aεsay of M2 substrate (with and without RJSTase treatment, lanes 1 and 2 respectively) , and synthetic telomerase products in heat-inactivated 293 extract were amplified in PCR asεay in both cold start (lane 3) and hot start (lanes 4-11) conditions. The amplified products were run on 15% polyacrylamide non-denaturing gel and exposed for 2 hr. on Phosphorlmager.

Fig. 37 is a schematic showing a PCR aεεay of this invention. Specifically, a one tube PCR-based telomerase aεεay is shown. The figure explains the formation of hot start PCR condition in the assay (1) , predicted telomerase reaction (2) , and the logic behind the telomerase product amplification and represεion of primer-dimer artifactε (3) . Fig. 38 iε a copy of autoradiograms showing results of asεays with extracts with various pretreatments. Specifically, the ability of telomerase to be active in PCR buffer condition (laneε 3 and 4) waε compared with the telomeraεe activity in conventional telomerase buffer. Conventional telomerase assays were performed in conventional and PCR buffer conditions. PCR-baεed telomerase assay in a single tube waε performed on active non-pretreated, and inactive pretreated 293 extractε (5-13) . Fig. 39 and 40 are copieε of autoradiogramε εhowing results of asεayε to measure εensitivity. Specifically, relative sensitivity was compared between conventional aεεay and PCR-based asεay. Active 293 extractε were diluted accordingly and used for conventional (lanes 2-5) and PCR-based (lanes 6-9) aεεayε. Conventional aεsayε were run with εynthetic telomerase product marker (M, lane 1) , exposed for 12 hr, and all of the reactions were loaded onto each lane. PCR-based aεεay waε expoεed for 2hr, half of the reactionε were loaded onto each lane.

PCR-baεed aεεayε were performed on different numberε of εynthetic telomeraεe product (M2+4, laneε 1-5) , dilutionε of concentrated extract (laneε 6-9) , normal fibroblaεt (lane 10) , and extractions made from different number of 293 cells (lanes 11-15) .

Fig. 41 shows resultε of telomeraεe assays on various cells and tisεues .

Fig. 42 is a graph showing terminal restriction fragment length (TRF) of blood cellε iεolated from AIDS patientε (CDA⁺, CD8⁺, Pbl) , age-marked controls (cont. 22-42y) and a normal centenarian (CEN Pbl) . Telomeres and Telomeraεe

All normal diploid vertebrate cells have a limited capacity to proliferate, a phenomenon that haε come to be known as the Hayflick limit or replicative senescence. In human fibroblasts, thiε limit occurε after 50-100 population doublingε, after which the cellε remain in a viable but non-dividing senescent state for many months . This contrastε to the behavior of most cancer cells, which have escaped from the controls limiting their proliferative capacity and are effectively immortal.

One hypothesis to explain the cause of cellular senescence concerns the role of the distal ends of chromosomes called telomeres . The hypothesis is that somatic cells do not express the enzyme telomerase and therefore lack the ability to replicate the very ends of DNA molecules. This results in a progressive shortening of the ends of the chromosomeε until εome function changeε, at which time the cell loεeε the capacity to proliferate. DNA polymeraεe synthesizes DNA in a 5 'to 3' direction and requires a primer to initiate syntheεis. Because of thiε, the "lagging strand" does not replicate to the very ends of linear chromosomes . The chromosome is thus shortened with every cell division. The ends of chromosomeε are called telomeres, and are composed of long TTAGGG repeatε . The enzyme telomerase can add TTAGGG repeats to the 3' end of the telomeric DNA, thuε extending the DNA and preventing εhortening. Germline cells are immortal, have long telomeres, and active telomerase. Somatic cells lack telomerase activity, and their telomeres have been found to shorten with cell division both in vivo and in culture. Cancer cells are immortal, and have regained telomeraεe activity and thuε can maintain their chromosome ends. Examples are provided below of definitive experiments which indicate that telomere shortening and telomerase activity are key factors in controlling cellular senescence and immortalization. Methods

As noted above, the present invention concernε diagnoεiε and therapy associated with measuring telomeric length and manipulating telomerase-dependent extension or telomerase-independent shortening. While the invention is directed to humans, it may be applied to other animalε, particularly mammals, such as other primates, and domestic animals, such aε equine, bovine, avian, ovine, porcine, feline, and canine. The invention may be uεed in both therapy and diagnosis. In this case of therapy, for example, telomere shortening may be slowed or inhibited by providing DNA oligonucleotides, by reactivating or introducing telomerase activity, or their functional equivalent, or indefinite proliferation can be reduced by inhibiting telomerase. In the case of diagnoεticε, one may detect the length of telomereε as to a particular chromosome or group of chromosomes, or the average length of telomeres. Diagnosis may also be associated with determining the activity of telomeraεe, or the preεenεe of the components of the enzyme either on a protein or RNA level, in cellε, tiεsue, and the like.

Information on the relative age, remaining proliferative capacity, as well as other cellular characteriεticε aεεociated with telomere and telomerase status may be obtained with a wide variety of cell typeε and tiεsues, such as embryonic cells, other stem cells, somatic cells (such as hepatocyteε in the context of cirrhoεiε) , connective tissue cells (such as fibroblaεts, chondrocytes, and osteoblastε) , vaεcular cellε (such as endothelial and smooth muscle cellε) , cellε located in the central nervouε εystem (such aε brain aεtrocyteε) , and different neoplaεtic tiεεueε, and paraεitic pathogens where it is desirable to determine both the remaining replicative capacity of the hyperplastic cells and their capacity for immortal growth to predict growth potential. Maintaining Telomere Length

Telomere length in cellε in vi tro or in vivo may be usefully maintained by a variety of procedures. These include those methods exemplified below. These examples, however, are not limiting in this invention εince those in the art will recognize equivalent methods. It is expected that all the methods will be useful in manipulating telomere length now that applicant haε demonstrated this experimentally. Such methods may be baεed upon proviεion of oligonucleotides or other agents which interact with telomereε to prevent shortening during cell division. In addition, the methods include treatment with agents which will include telomerase, or its equivalent activity, within a cell to prevent εhortening or extend telomeres. Finally, the methods also include modulation of gene expression associated with cell senescence.

Uεeful agents can be determined by routine screening procedures. For example, by screening agents which interact in an in vi tro εyεtem with telomeres, and block loss of telomere ends, or aid increaεe in telomere length. Non-limiting examples of such methodε are provided below. All that is neceεεary iε an aεsay to determine whether telomere end shortening iε reduced during cell diviεion. The mechaniεm by which εuch agents act need not be known, so long aε the deεired outcome is achieved. However, by identifying useful target genes (e.g.. the M2 mortality modulation gene(s)), antisense and equivalent procedures can be designed to more appropriately cause deεired gene expreεεion or non-expreεεion (e.g.. the de-repreεεion of telomeraεe) .

In a particular example (non-limiting in thiε invention) one can reduce the rate of telomere εhortening, by providing a nucleic acid, e.g. , DNA or RNA (including modified forms) , as a primer to the cells. Such nucleic acid will usually include 2 to 3 repeats, more usually 2 repeats, where the repeatε are complementary to the G-rich DNA telomere εtrand. Such oligonucleotides may be used to extend the proliferative capability of cells.

The oligonucleotideε can be transferred into the cytoplasm, either spontaneously (i.e., without specific modification) or by the use of liposomes which fuse with the cellular membrane, or are endocytosed by employing ligands which bind to εurface membrane protein receptorε of the cell reεulting in endocytoεiε. Alternatively, the cellε may be permeabilized to enhance tranεport of the oligonucleotideε into the cell, without injuring the hoεt cellε. Another way iε to uεe a DNA binding protein, e.g. , HBGF-1, which iε known to tranεport an oligonucleotide into a cell. In this manner, one may substantially reduce the rate of telomere εhortening from an average of about 50 bp per diviεion, to an average of about 6-12 bp per diviεion (εee exampleε below) , thuε significantly extending the number of divisions occurring before induced cellular senescence.

By "senescence" is meant the loss of ability of a cell to replicate in the preεence of normally appropriate replicative signals, and may be associated with the expresεion of degradative enzymes, such as collagenase. The term does not include quiescent cells which might be induced to replicate under appropriate conditions. Thiε term iε exemplified below in the examples, where the number of cell doubling prior to senescence is increased. The above processes are useful in vivo . As already indicated, by using liposomes, particularly where the liposome surface carries ligands εpecific for target cellε, or the lipoεomes will be preferentially directed to a specific organ, one may provide for the introduction of the oligonucleotides into the target cells in vivo . For instance, utilizing lipocortin affinity for phosphatidyl serine, which is releaεed from injured vaεcular endothelial cellε, the oligonucleotideε may be directed to εuch εite. Alternatively, catheters, syringeε, depotε or the like may be uεed to provide high localized concentrationε. The introduction of εuch oligonucleotides into cells resulting in decreased εeneεcence in reεponεe to cell division can have therapeutic effect. The maintenance of telomere length has application in tissue culture techniques to delay the onset of cellular εeneεcence. For inεtance, cell-baεed therapies which require the clonal expansion of cells for reintroduction into an autologous patient are limited to about 20-30 doublingε. This invention allows, the expansion of cells in the case of gene therapy, both prior to genetic manipulation and then expansion of the manipulated cells, the maintenance of telomere length. This in turn allows normal cells to be cultivated for extended doublings in vi tro . Experiments described below demonstrate the utility of thiε method in vi tro, and demonstrate its applicability in vivo .

Critical shortening of telomeres leads to a phenomenon termed "crisiε" or M2 εeneεcence. See, Shay et al. , 1992, supra . Among the cellε in crisis, rare mutants may become immortalized in which M2 genes have altered regulation, and where expreεεion of telomeraεe is reactivated and stabilizes the telomere length. An M2 regulatory gene may be modulated to provide a uεeful means of modulating telomere length and telomerase activity. The M2 geneε may be identified by means of insertional mutagenesis into cells in M2 crisis utilizing a retrovirus. Cellε wherein the M2 gene haε been knocked out will then grow in response to the re¬ activation of telomerase, and such cellε can supply a source or DNA from which to clone the M2 genes. This technique has yielded numerous cell clones in which the retrovirus has inserted into a common restriction fragment. The repression of the M2 regulatory gene(s) by antisenεe or other means can provide a means of activating telomeraεe reverεibly, εuch that telomeres may be extended and then telomerase again represεed. In thiε manner, proliferative capacity may be extended with or without the addition of oligonucleotideε to εlow the telomere shortening. Such cells may then be used in cell-based therapieε, εuch as bone marrow transplantation, reconstitution of connective tiεsue, and transplantation of early passage adrenal cortical cells, fibroblasts, epithelial cells, and myoblasts.

Increaεed replicative capacity may be imparted to cultured cellε by meanε of the tranεient introduction of telomeraεe activity.

Telomeraεe can be iεolated from immortal human cellε for use in these procedures. Telomerase may be purified by extraction in either hypotonic buffer or non-ionic detergent. It can also be purified by passing over a DEAE column and subεequent purification techniqueε. Telomerase can then be reintroduced to cells either by lipoεome mediated addition or by micro- injection. The εource of cellε containing telomeraεe would be the human tumor cell line εuch as U937 histiocytic lymphoma.

Telomeraεe can also be isolated from altered Tetrahymena. Tetrahymena synthesizes a telomere repeat of 5' TTGGGG 3' . The template on an encoding sequence is cloned and can be altered in the sequence to encode the human telomere repeat 5' TTAGGG 3' . The tetrahymena enzyme may then be reconεtituted with the altered RNA sequence to produce telomerase enzymes synthesizing the human telomeric sequence. This enzyme can be obtained in large quantities from Tetrahymena, purified arid added to cells. Recombinant telomerase may be produced in highly purified form once the telomerase cDNA and template RNA are cloned.

The C-rich terminal repeat mRNA may be expreεεed in cellε in parallel with the expreεεion of a reverεe tranεcriptase activity from, for instance, HIV. The reverεe tranεcriptaεe activity can be imparted either by transfection of cDNA or lipoεome mediated delivery of protein. The reεulting combination is expected to have a telomerase activity with the CTR mRNA forming the template for reverse transcription. Such a construct can be added to cells using presently existing technology.

Reactivation of repressed telomerase, activity may be posεible once agentε are found that may induce the enzyme. Such agentε may be identified utilizing εcreening technologieε deεcribed herein. Reactivation of repreεεed telomeraεe activity by agentε identified as described herein also haε important therapeutic applicationε. Meanε of delivery of telomeraεe to cells may include liposome mediated addition or micro-injection.

In addition, telomerase activity may be added to cells by meanε gene therapy uεing vehicles to transport the mRNAs for the telomeraεe componentε, or the genes for the components into cells.

Telomerase can be used in many different tissue and cell types. For example, telomerase may be useful when applied directly to the dermis. It is possible that the replicative seneεcence of dermal fibroblaεtε iε reεponsible for the poor wound healing observed in the elderly. Theεe individuals often experience chronic nonhealing skin leεionε such as staεiε ulcerε and decubituε ulcerε. Telomeraεe can be applied directly to the wound to increaεe the replicative capacity of fibroblaεtε and keratinocyteε in the wound. The technique iε alεo useful in cases of burns covering large areas of skin, where the repopulation of the surface area would require cells to replicate to the end of their capacity. Similarly, where attempts are made to aid in healing of large burns using skin synthesized in vitro, replicative seneεcence may limit the ability to regenerate skin, and means to increase the replicative capacity of the cells would be useful. It is also useful to inject telomerase locally into regions where it is desirable to decreaεe the expreεsion of genes associated with telomere repeat losε, aε in εkin wrinkles.

The endothelium is unique in that it iε easily accesεible via the blood. The administration of telomerase activity to aged endothelial cellε may increase their replicative capacity thereby promoting the covering of the lesions frequently not covered in serile lesions. The addition of telomeric repeats by use of telomerase may also down-regulate the expression of senescent-specific genes.

The addition of telomeric repeatε to aging brain aεtrocytes and endothelial cells would be expected to allow the cells to exit the cell cycle in the normal Go state thereby down-regulating the expreεεion of amyloidogenic proteinε cauεative in Alzheimer'ε diεeaεe. The aging eye iε characterized by εpecific changeε in the retina in aεεociation with a layer of cellε called retinal pigmented epithelial (RPE) cellε. In the region of the retina called the macula, these cells are expoεed to high levels of damaging UV radiation and therefore are supplied with regenerative capacity for repair. In the aging eye, degenerative changes occur in asεociation with the RPE layer. The healthy retina iε avaεcular. The RPE εecreteε factorε that inhibit angiogeneεiε. The RPE alεo secretes factors that effect the differentiative function of the retinal neurons. RPE cells can be taken from the periphery of the retina of an individual where there has been minimal UV damage, the cells selected and/or expanded in the presence of CTR (aε deεcribed infra) , or transiently treated with telomerase and reintroduced into the same individual. The transient administration of telomeraεe activity to the RPE may down-regulate the expreεsion of seneεcent-εpecific gene expreεsion and thereby provide a uεeful therapeutic approach.

The εeneεcence of chondrocyteε leads to the overexpresεion of the destructive proteins collagenase and εtromelyεin that destroy articular cartilage in osteoarthritiε. Strategies to transiently expreεε telomeraεe in aging chondrocyteε will also have therapeutic effect to increase the replicative capacity of the chondrocyteε and down-regulate εeneεcent gene expression.

In some cell types it may be beneficial to expresε telomerase transiently in order not to permanently immortalize the cell. In some cells immortalization may prediεpoεe the cell to transforming into a malignant tumor cell. The transient expresεion of telomeraεe along with factorε that increaεe the processivity of the enzyme (such as the GTO oligonucleotide shown in Fig. 15) may be sufficient to greatly extend the replicative capacity of the cells without permanent immortalization.

Telomerase Modulation As discuεεed above, cancer cellε contain telomeraεe activity and are thereby immortal. In addition, numerous types of parasitic pathogens are immortal and have active telomerase. Thuε, it iε useful to modulate (e.g. , decreaεe) telomeraεe activity in such cells to impart a finite replicative life span. In contrast to the long telomeric tracts in normal human cells, tracts of telomeric DNA in protozoan cells, fungal cells, and some parasitic worms, aε well aε many cancer cells, are typically shorter. This makes these cells more vulnerable to telomerase inhibitors than normal human cells (e.g. germ line cells) .

Thus, inhibition or induction of telomerase has applications in various situationε. By inhibiting telomerase intracellularly, one may reduce the ability of cancer cells to proliferate. Telomerase may be competitively inhibited by adding εynthetic agents, e.g. , oligonucleotides comprising 2 or more, usually not more than about 50 repeats, of the telomeric motif of the 5' -3' G-rich strand (the strand which acts aε the template) . The oligonucleotideε may be εyntheεized from natural or unnatural unitε, e.g. , the derivativeε or carbon derivativeε, where a phoεphate-oxygen is substituted with sulfur or methylene, modified sugarε, e.g. , arabinoεe, or the like. Aε diεcussed above, other equivalent agents may also be used to inhibit or cause expresεion of telomerase activity.

The oligonucleotides may be introduced as described above so as to induce seneεcence in the immortalized cellε, in culture and in vivo . Where growing cellε in culture, where one wiεheε to prevent immortalized cellε from overgrowing the culture, one may use the subject oligonucleotides to reduce the probability of such overgrowth. Thus, by maintaining the oligonucleotides in the medium, they will be taken up by the cells and inhibit telomeraεe activity. One may provide for linkage to the telomeric sequence with a metal chelate, which results in cleavage of nucleic acid sequences. Thus, by providing iron chelate bound to the telomeric motif, the telomerase RNA will be cleaved, so as to be non-functional. Alternatively, a reactive group may be coupled to the oligonucleotide that will covalently bind to telomerase, or the 3' reεidue may be made to be dideoxy εo aε to force chain termination. Alternatively, one may introduce a ribozyme, having 5' and 3'-terminal εequences complementary to the telomerase RNA, so as to provide for cleavage of the RNA. In this way, the telomerase activity may be εubεtantially inhibited, so as to result in a significant limitation of the ability of the cancer cells to proliferate.

Ribozymes are RNA moleculeε having an enzymatic activity which is able to repeatedly cleave other separate RNA moleculeε in a nucleotide baεe εequence εpecific manner. Such enzymatic RNA moleculeε can be targeted to virtually any RNA tranεcript, and efficient cleavage haε been achieved in vi tro . Kim et al., 84 Proc. Natl. Acad. Sci. USA 8788, 1987; Haseloff and Gerlach, 334 Nature 585, 1988; Cech, 260 JAMA 3030, 1988; and Jefferies et al. , 17 Nucleic Acidε Research 1371, 1989.

Ribozymes act by first binding to a target RNA. Such binding occurs through the target RNA binding portion of a ribozyme which is held in close proximity to an enzymatic portion of the RNA which acts to cleave the target RNA. Thus, the ribozyme first recognizes and then binds a target RNA through complementary base-pairing, and once bound to the correct site, acts enzymatically to cut the target RNA. Strategic cleavage of εuch a target RNA will deεtroy its ability to direct syntheεis of an encoded protein. After a ribozyme has bound and cleaved itε RNA target it is released from that RNA to search for another target and can repeatedly bind and cleave new targetε. The enzymatic nature of a ribozyme is advantageous over other technologies, such as antisense technology (where a nucleic acid molecule simply binds to a nucleic acid target to block its translation) since the effective concentration of ribozyme necessary to effect a therapeutic treatment iε lower than that of an antiεense oligonucleotide. Thiε advantage reflectε the ability of the ribozyme to act enzymatically. Thus, a single ribozyme molecule is able to cleave many molecules of target RNA. In addition, the ribozyme is a highly specific inhibitor, with the specificity of inhibition depending not only on the base pairing mechanism of binding, but also on the mechanism by which the molecule inhibits the expression of the RNA to which it binds. That is, the inhibition is caused by cleavage of the RNA target and so specificity is defined as the ratio of the rate of cleavage of the targeted RNA over the rate of cleavage of non-targeted RNA. This cleavage mechanism is dependent upon factors additional to those involved in baεe pairing. Thuε, it iε thought that the specificity of action of a ribozyme is greater than that of antisense oligonucleotide binding the same RNA site.

A ribozyme is an "enzymatic RNA molecule" in that it is an RNA molecule which has complementarity in a substrate binding region to a specified gene target, and also has an enzymatic activity which is active to specifically cleave RNA in that target. That is, the enzymatic RNA molecule is able to intermolecularly cleave RNA and thereby inactivate a target RNA molecule. This complementarity functions to allow sufficient hybridization of the enzymatic RNA molecule to the target RNA to allow the cleavage to occur. One hundred percent complementarity is preferred, but complementarity as low aε 50-75% may alεo be uεeful. Ribozymes targeting any of the specific telomerase coding regions deεcribed in detail herein should be able to cleave the RNAs in a manner which will inhibit the tranεlation of the moleculeε and thuε reduce telomeraεe activity. In addition, ribozymeε targeting the naεcent RNA guide εequence of the telomeraεe will reduce telomeraεe activity.

In preferred embodimentε, the enzymatic RNA molecule iε formed in a hammerhead motif, but may also be formed in the motif of a hairpin, hepatitis delta virus, group I intron or RNaseP-like RNA (in aεεociation with an RNA guide εequence) . Exampleε of εuch hammerhead motifε are deεcribed by Rossi et al. , 8 Aids Reεearch and Human Retroviruεes 183, 1992; of hairpin motifs by Hampel and Tritz, 28 Biochemistry 4929, 1989 and Hampel et al. , 18 Nucleic Acids Research 299, 1990; an example of the hepatitis delta virus motif is described by Perrotta and Been, 31 Biochemistry 16, 1992; of the RNaseP motif by Guerrier- Takada et al. , 35 Cell 849, 1983; and of the group I intron by Cech et al., U.S. Patent 4,987,071. These specific motifε are not limiting in the invention and thoεe εkilled in the art will recognize that all that iε important in an enzymatic RNA molecule of thiε invention iε that it haε a εpecific εubεtrate binding εite which iε complementary to one or more of the target gene RNA regionε, and that it have nucleotide εequenceε within or εurrounding that substrate binding site which impart an RNA cleaving activity to the molecule.

The ε alleεt ribozyme delivered for treatment of HIV infection reported to date (by Roεεi et al. , 1992, supra) is an in vi tro transcript having a length of 142 nucleotides. Synthesis of ribozymeε greater than 100 nucleotideε in length is very difficult uεing automated methodε, and the therapeutic coεt of εuch moleculeε iε prohibitive. Delivery of ribozymes by expression vectors is primarily feasible using only ex vivo treatments. This limits the utility of this approach. In this invention, small ribozyme motifs (e.g. , of the hammerhead structure, or of the hairpin structure) are used for exogenous delivery. The simple structure of these molecules also increases the ability of the ribozyme to invade targeted regions of the mRNA structure.

One potential telomerase RNA target for a ribozyme has the sequence 3' AUCCCAAUC 5' which is a portion of the nascent RNA required for telomerase activity. Other potential targets may be determined by reviewing the RNA sequence of the nascent RNA, or of an mRNA encoding telomerase, as noted above. Telomeraεe may alεo be inhibited by the adminiεtration of an M2 regulator gene product. By modulating the expression of any of the proteins directly regulating telomerase expression, one may also modulate cellular telomerase activity. Alternatively, one may use a screening assay utilizing human or tetrahymena telomerase to screen small molecules e.g., nucleoside analogε like ara-G, ddG, AZT, and the like and RNA and DNA processing enzyme inhibitors, alkylating agentε, and various potential anti-tumor drugs. These may then be further modified.

The nucleic acid sequences may be introduced into the cells as described previously. Various techniques exist to allow for depots associated with tumors. Thus, the inhibiting agents or nucleic acids may be administered aε drugε, εince they will only be effective only in cells which include telomerase. Since for the most part, human somatic cells lack telomerase activity, they will be unaffected. Some care may be required to prevent entry of εuch drugε into germ cellε or εome εtem cell populations, which may express telomerase activity. The subject compositionε can therefore be used in the treatment of neoplasia wherein the tumor cells have acquired an immortal phenotype through the inappropriate activation of telomerase, as well as various human and veterinary parasitic diseases; including human protozoal pathogens such as; amebiasis from Entamoeba histolytica, amebic meningoencephalitiε from the genuε Naegleria or Acanthamoeba, malaria from Plasmodium vivax, Plasmodium ovale, Plasmodium malariae , and Plasmodium falciparum, leiεhmaniasis from εuch protozoa aε Leishmania donovani , Leishmania infantum, Leishmania chagasi , Leishmania tropica, Leishmania major, Leishmania aethiopica, Leishmania mexicana, and Leishmania braziliensis, Chagaε' disease from the protozoan Trypanosoma cruzi , sleeping sicknesε from Trypanosoma brucei , Trypanosoma gambiense, and Trypanosoma rhodesiense, toxoplasmosiε from Toxoplasma gondii , giardiaεiε from Giardia lamblia , cryptoεporidioεiε from Crypto sporidium parvum, trichomoniaεiε from Trichomonas vaginalis, Trichomonaε tenax, Trichomonas hominis, pneumocystis pneumonia from Pneumocystis carinii , bambeεoεiε from Bambesia microti , Bambesia divergens , and Bambesia boriε, and other protozoanε causing intestinal diεorderε such as Balantidium coli and Isospora belli . Telomerase inhibitorε would alεo be uεeful in treating certain helminthic infectionε including the εpecieε: Taenia solium, Taenia saginata, Diphyllobothrium lata,

Echinococcus granulosus , Echinococcus mul tilocularis , Hymenolepis nana, Schistosoma mansomi , Schistosoma japonicum, Schistosoma hematobium, Clonorchis sinensis, Paragonimus westermani , Fasciola hepatica, Fasciolopsis buski , Heterophyes heterophyes , Enterobi us vermi cular is , Trichuris tri chi ura , As car is l umbri coides , Ancylostoma duodenale, Necator americanus, Strongyloides stercoraliε, Tri chinella spiralis , Wuchereria bancrofti , Onchocerca volvulus, Loa loa, Dracunculus medinensis, and fungal pathogens such as: Sporothrix schenckii , Coccidioides immitis, Histoplasma capsulatum, Blastomyces dermatitidis, Paracoccidioides brasiliensis, Candida albicans, Cryptococcus neoformans, Aspergilluε fumigatus , Aspergillus flavuε, fungi of the genera Mucor and Rhizopus, and εpecies cauεing chromomycoεis such as those of the genera Phialophora and Cladosporium . and important veterinary protozoal pathogens such as: Babesia caballi, Babesia caniε, Babesia equi , Babesia felis, Balantidium coli , Besnoitia darling i , Eimeria acervulina, Eimeria adenoeideε , Eimeria ahsata, Eimeria alabamensis, Eimeria auburnenεiε, Eimeria bovis, Eimeria brasiliensiε, Eimeria brunetti, Eimeria canadensis, Eimeria cerdonis, Eimeria crandalliε, Eimeria cylindrica, Eimeria debliecki, Eimeria deεpersa, Eimeria ellipεoidalis , Eimeria fauvei, Eimeria gallopavoni , Eimeria gilruthi, Eimeria granulosa, Eimeria hagani , Eimeria illinoisensis, Eimeria innocua, Eimeria intricata, Eimeria leuεkarti , Eimeria maxima, Eimeria meleagridiε, Eimeria meleagrimitis, Eimeria mitis, Eimeria mivati , Eimeria neca trix, Eimeri a neodebl i ecki , Eimeri a ninakohlyakimorae , Eimeria ovina, Eimeria pallida, Eimeria parva, Eimeria perminuta, Eimeria porci , Eimeria praecox, Eimeria punctata, Eimeria εcabra, Eimeria εpinoza, Eimeria εubrotunda, Eimeria εubεherica, Eimeria suis, Eimeria tenella, Eimeria wyomingensi ε , Eimeri a zuerni i , Endol imax gregariniformiε, Endolimax nana, Entamoeba bovis, Entamoeba gallinarum, Entamoeba hiεtolytica, Entamoeba suis, Giardia bovis, Giardia canis, Giardia cati , Giardia Iambi ia, Haemoproteus meleagridiε, Hexamita meleagridis, Hiεtomonaε meleagridis, Iodamoeba buetschili , Iεoεpora bahiensiε, Iεoεpora burrowεi , Isospora caniε, Iεoεpora feliε, Iεoεpora ohioenεiε, Isospora rivolta, Isospora suis, Klossiella equi , Leucocytozoon caallergi, Leucocytozoon εmithi ,

Parahiεtomonas wenrichi, Pentatrichomonas hominiε, Sarcocystis betrami , Sarcocystis bigemina, Sarcocystiε cruzi , Sarcocyεtiε fayevi , hemionilatrantiε, Sarcocystiε hirsuta, Sarcocystiε mieεcheviana,

Sarcocyεtiε muris, Sarcocyεtiε ovicaniε, Sarcocyεtiε tenella, Tetratrichomonaε buttreyi, Tetratrichomonaε gallinarum, Theileria mutanε, Toxoplaεma gondii , Toxoplasma hammondi , Trichomonas caniεtomae, Trichomonas gallinae, Trichomonas feliεtomae,

Trichomonaε eberthi, Trichomonaε equi , Trichomonaε foetuε, Trichomonaε oviε, Trichomonaε rotunda, Trichomonas suis, and Trypanosoma melophagium. In addition, they can be uεed for εtudying cell senescence, the role of telomeres in the differentiation and maturation of cells from a totipotent stem cell, e.g. , embryonic stem cells, or the like, and the role of telomerase in spermatogeneεis . Telomere Length

Procedures for measuring telomere length are known in the art and can be used in this invention.

Typically, restriction endonuclease digeεtion is used

(with enzymes which do not cleave telomeric DNA) , and the length of the fragment having detectable telomere DNA iε εeparated according to molecular weight by agaroεe gel electrophoresis . Given that the DNA sequence of a telomere is known, detection of εuch DNA is relatively easy by use of specific oligonucleotides. Examples of these methods are provided below.

For diagnosiε, in detection of the telomeric length, one may εtudy just a particular cell type, all cells in a tisεue (where variouε cellε may be preεent) , or subsetε of cell types, and the like. The preparation of the DNA having such telomereε may be varied, depending upon how the telomeric length is to be determined. Conveniently, the DNA may be iεolated in accordance with any conventional manner, freeing the DNA of proteinε by extraction, followed by precipitation. Whole genomic DNA may then be melted by heating to at least about 80°C, usually at least about 94°C, or using high salt content with chaotropic ions, such as 6X SSC, quanidinium thiocyanate, urea, and the like. Depending upon the nature of the melting procesε, the medium may then be changed to a medium which allows for DNA εynthesis.

(a) DNA Synthesis

In one method, a primer iε used having at least about 2 repeats, preferably at least about 3 repeatε of the telomeric sequence, generally not more than about 8 repeats, conveniently not more than about 6 repeats . The primer is added to the genomic DNA in the presence of only 3 of the 4 nucleoside triphosphateε (having the complementary nucleoεides to the protruding or G-rich strand of a telomere, e.g.. A, T and C for human chromosomes), dATP, dTTP and dCTP. Usually at least the primer or at least one of the triphosphates iε labeled with a detectable label, e.g. , a radioiεotope, which label iε retained upon incorporation in the chain. If no label iε used, other methods can be used to detect DNA synthesis. The primer is extended by means of a DNA polymerase, e.g.. the Klenow fragment of DNA polymerase I, T7 DNA polymerase or Taq DNA polymerase

The length of the extended DNA can then be determined by various techniques, e.g. , those which separate synthesized DNA on the basis of its molecular weight, e.g. , gel electrophoresis. The DNA synthesized may then be detected based on the label, e.g. , counts incorporated per μg of DNA, where the counts will be directly proportional to telomere length. Thuε, the measure of radioactivity in relation to the amount of DNA will suffice to quantitate telomere length.

If desired, telomeres of known length may be used as standards, whereby a determination of radioactivity may be read off a standard curve as related to telomere length. Instead, one may prepare tisεueε where individual cells may be asεayed for relative telomere length by in εi tu hybridization. In this approach, for example, the primer is labeled with a detectable label, usually biotin or digoxygenin. Following annealing to prepared tisεue sections or cellε, the label iε revealed histochemically, usually using autoradiography (if the label were radioactive) , uεing avidin/streptavidin (if the label were biotin) or using antidigoxygenin antibodies (if the label were digoxygenin) . The amount of signal per cell is proportional to the number of telomeric repeats, and thus to the telomere length. This can be quantitated by microfluorometry or analogous means, and compared to the signal from standard cells of known telomere length to determine the telomere length in the teεt sample.

(b) Restriction Endonuclease Digestion Alternatively, one may use primers which cause covalent crosε-linking of the primer to telomere DNA. In thiε situation, one may totally digeεt the DNA with reεtriction endonucleaεeε which have 4 baεe recognition εiteε, which reεults in the production of relatively short fragments of DNA, except for telomeric DNA which lacks the recognition site. Reεtriction endonucleaεeε which may find uεe include Alul, Hinfl , Mεpl , Rεal , and Sau3A, where the reεtriction endonucleaεes may be used individually or in combination. After digestion of the genomic DNA, the primer may be added under hybridizing conditions, so as to bind to the protruding chain of the telomeric sequence. By providing for two moieties bound to the primer, one for covalent bonding to the telomeric sequence and the other for complex formation with a specific binding pair member, one can then provide for linking of a telomeric sequence to a surface. For example, for covalent bonding to the telomeric sequence, psoralen, or iεopsoralen, may be linked to one of the nucleotides by a bond or chain and upon UV-radiation, will form a bridge between the primer and the telomere.

The specific binding pair member will normally be a hapten, which binds to an appropriate complementary member, e.g.. biotin and strept/avidin, trinitrobenzoic acid and anti-trinitrobenzamide antibody, or methotrexate and dihydrofolate reductase. Rather than having the moiety for covalent bonding covalently bonded to the primer, one may add a compound into the medium which is intercalatable into the nucleic acid, εo aε to intercalate between double- εtranded nucleic acid εequences. In this manner, one may achieve the same purpoεe. Uεe of a subεtantial excess of the intercalatable compound will cause it to also intercalate into other portions of DNA which are present. Various modifications of this proceεε may be achieved, εuch aε εize εeparation, to reduce the amount of label containing DNA.

The εpecific binding pair member may be uεed for εeparation of telomeric DNA free of contaminating DNA by binding to the complementary pair member, which may be present on beads, on particles in a column, or the like. In accordance with the nature of the separation, the covalently bonded telomere strand may now be purified and measured for εize or molecular weight. Again, if desired, standardε may be employed for comparison of distribution values.

The specific binding pair member hapten can be present at the 5' -terminus of the primer or at intermediate nucleotideε. Specifically, biotin- conjugated nucleotideε are generally available and may be readily introduced into εynthetic primer εequenceε in accordance with known wayε. The above-deεcribed techniqueε can alεo be uεed for isolating and identifying DNA contiguous to the telomere.

(c) Average Telomere Length In methods of this invention it may be useful to determine average telomere length by binding a primer to a telomere prior to separation of the telomeric portion of the chromosomes from other parts of the chromosomes. This provides a double-stranded telomeric DNA comprising the telomeric overhang and the primer. A reaction may then be carried out which allowε for εpecific identification of the telomeric DNA, aε compared to the other DNA preεent. The reaction may involve extenεion of the primer with only 3 of the nucleotideε (dNTPε) , uεing a labeled nucleotide, covalent bonding of the primer to the telomeric εequence, or other methodε which allow for separation of the telomeric sequence from other sequences . The length of the synthesized DNA detected then representε the average telomere length.

Telomere length can alεo be meaεured directly by the "anchored terminal primer" method. In thiε method, the 3' endε of genomic DNA are first "tailed" with dG nucleotides using terminal transferase. Telomereε, which are known to have 3' overhangε, then would have one of the three follwing conformationε: 5'TTAGGGTTAGGGTTAGGGGGGGGGGG...3' 5'TTAGGGTTAGGGTTGGGGGGGGGGGG...3' 5'TTAGGGTTAGGGTGGGGGGGGGGGGG...3'

Other ends of the genomic DNA which were generated by shearing would be tailed with G's but would not have the adjacent TTAGGG repeatε. Thuε, a mix of the following 3 biotinylated oligonucleotides would anneal under stringent conditions specifically to all posεible telomere ends :

5'B-CCCCCCCCTAACCCTA

5'B-CCCCCCCCAACCCTAA Oligo Mix [M] 5'B-CCCCCCCCACCCTAAC Oligo mix [M] consists of 16-base oligonucleotideε with 5' biotin (B) , but other combinationε of 5'-C-tracts adjacent to the C-rich telomeric repeats could provide specific hybridization to the 3' end of the native telomeres.

Extension of the primer with a DNA polymerase such as Klenow, DNA Polymerase I, or Taq polymerase, in the preεence of dCTP, dATP, dTTP (no dGTP, and with or without ddGTP) would stabilize the primer-template configuration and allow selection, using streptavadin beads, of the terminal fragments of DNA containing the telomeric DNA. The length of primer extension using Klenow (monitored with labeled nucleotides) would indicate the length of the telomeric (GTR) 3' overhang, since Klenow lacks 5' -3' exonuclease activity and would stall at the CTR. This length distribution could be indicative of the level of telomerase activity in telomerase-positive cellε (i.e., longer extenεions correspond to greater telomerase activity) . In contrast, extension of the primer with DNA polymerase

I, an enzyme with 5' -3' exonuclease activity as well as polymerase activity, would allow extension through the

CTR until C's are encountered in the template εtrand

(εubtelomeric to the GTR) . The length distribution of this reaction, monitored by labeled nucleotides, would be indicative of the length distribution of the GTR. In both cases, labeled products arising from biotinylated primers are selected with the streptavadin beads to reduce the signal from non-specific priming. Alternatively, re-priming and extenεion of the tailed chromosome end can take place after selection of the partially extended products with the streptavadin beads, and after denaturation of the C-rich strand from the duplex. Experiments have confirmed that the G-tailing of chromosome ends can be carried out efficiently such that about 50 G residues are added per end, that the priming with the junction oligonucleotide mix is highly specific for the tailed telomeric ends, and that streptavadin beads select εpecifically for the extension products that originate from the biotinylated primers and not from other fortuitous priming events. The length of the extenεion productε under the conditionε outlined above thuε provide a direct estimate of the length of the terminal TTAGGG repeat tract. This information is especially important in cases where stretches of TTAGGG repeats occur close to but not at the termini of chromosomeε. No other method described to date is capable of distinguiεhing between the truly terminal TTAGGG repeats and such internal repeatε. It iε poεεible to determine the amount of telomeric DNA on individual chromosomes by FISH using fluoreεcently labeled oligo- or polynucleotide probes. Chromoεomeε can be collected from metaphaεe cells, wherein they are identified by shape and/or banding patterns using staining procedures or secondary probes of a different fluorescent color, or they can be spread and stretched from interphaεe cellε. In the later caεe, it iε poεεible again to identify εpecific chromosomes with fluoreεcently labeled εecondary probes complementary to sequences cloεe to the telomere. Quantitative FISH with confocal microεcopy or imaging systems uεing εignal integration or contour length allowε one to obtain an objective meaεure of the distribution of telomere lengths on different chromoεomeε and to identify chromoεomeε which have potentially loεt a critical amount of telomeric DNA.

The determination of telomere length aε deεcribed above can be aεεociated with a variety of conditionε of diagnoεtic intereεt. Following telomere length in tumor cellε provides information regarding the proliferative capacity of such cellε before and following adminiεtration of inhibitors of telomerase (or other treatments which destabilizeε the telomere length aε diεcussed above) . It also provides a means of following the efficacy of any treatment and providing a prognosis of the course of the disease. Where diseased tisεue is involved, the native tissue can be evaluated as to proliferative capability. By "proliferative capability" is meant the inherent ability of a cell or cells in a tissue to divide for a fixed number of divisions under normal proliferation conditions. That is, the "Hayflick" number of divisions, exemplified below in the examples. Thus, despite the fact that the tissue may have a spectrum of cells of different proliferative capability, the average value will be informative of the state of the tissue generally. One may take a biopsy of the tisεue and determine the average telomeric length. Using the value, one may then compare the value to average normal healthy tisεue as to proliferative capability, particularly where the tissue is compared to other tiεsue of εimilar age.

In caεes of cellular diεeases, such as liver diseaεe, e.g. , cirrhoεis, or muscle diseaεe, e.g. , muεcular dystrophy, knowledge of the proliferative capability can be useful in diagnosing the . likely recuperative capability of the patient. Other situations involve injury to tisεue, such as in surgery, wounds, burns, and the like, where the ability of fibroblaεtε to regenerate the tiεεue will be of intereεt. Similarly, in the case of loss of bone, oεteoarthritiε, or other diεeases requiring reformation of bone, renewal capability of osteoblasts and chondrocytes will be of intereεt.

While methodε are deεcribed herein to evaluate the proliferative capacity of a tiεεue by taking an average meaεure of telomere length it iε noted that the tiεsue may have a spectrum of cells of different proliferative capability. Indeed, many tissueε, including liver, regenerate from only a small number of stem cells (lesε than a few percent of total cells) . Therefore, it iε uεeful in thiε invention to uεe in εi tu hybridization (εuch aε with fluoreεcently labeled telomeric probeε) , to identify and quantitate such stem cells, and/or the telomeric εtatus of such cells on an individual, rather than collective basiε. Thiε iε performed by meaεuring the fluoreεcent intensity for each individual cell nucleus using, e.g.. automated microscopy imaging apparatus. In addition to in si tu hybridization, gel electrophoresis is useful in conjunction with autoradiography to determine not only the average telomere length in cells in a tissue sample, but also the longeεt telomere lengths (posεibly indicating the preεence of stem cells) and the size distribution of telomere lengths (which may reflect different histological cell types within a tiεεue, see Figs. 10-11) . Thuε, the autoradiogram, or itε equivalent provideε useful information as to the total telomere εtatus of a cell, or group of cells. Each segment of εuch information is useful in diagnostic procedureε of thiε invention. d) Modified Maxam-Gilbert Reaction

The moεt common technique currently uεed to meaεure telomere length iε to digeεt the genomic DNA with a reεtriction enzyme with a four-baεe recognition εequence like Hinfl, electrophorese the DNA and perform a Southern blot hybridizing the DNA to a radiolabeled (TTAGGG)₃ probe. A difficulty with this technique is that the resulting terminal restriction fragments (TRFs) contain a 3-5 kbp stretch of εubtelomeric DNA that lackε restriction sites and thereby adds significantly to the εize of the meaεured telomere length. Another approach to eliminate this DNA and improve accuracy of telomere length asεayε utilizeε the fact that thiε εubtelomeric DNA containε G and C residues in both strandε, and thuε should be cleaved under conditionε that cause breaks at G residues. In contrast, DNA composed exclusively of telomeric repeats will have one strand lacking G residues, and this strand εhould remain intact under G- cleavage conditions. The Maxam-Gilbert G-reaction uses piperidine to cleave guanine residies that have been methylated by dimethylsulfate (DMS) treatment. Although the original conditions of the Maxam-Gilbert G-reaction (treatment in IM piperidine for 30 min. at 90° C) breaks unmethylated DNA into fragments of 1-2 kbp and is thus non-specific, milder conditions (0.1M piperidine for 30 min. at 37^*C) leave untreated DNA intact. The DNA is therefore treated with DMS and piperidine as described above, precipitated with ethanol, electrophoresed, and hybridized on a Southern blot to the a (TTAGGG)₃ probe. The results of such a test are shown in Fig. 26. Telomerase Activity

Telomerase activity has been detected in cell-free extracts of dividing, cultured hematopoietic εtem or early progenitor cells but not other more differentiated dividing cells. Thus, telomerase activity and or molecular probes, such as antibodieε or cDNA, may be used to distinguiεh certain stem cells or early progenitor cellε from more differentiated cellε which lack telomeraεe. Such probeε may allow one to εelect, by FACS or equivalent methodε, cellε having high proliferative and/or self-renewal capacity and posεibly a pluripotent ability for differentiation. The ability to εelect εtem and/or early progenitor cellε iε important for maximizing growth and differentiation during ex-vivo expanεion of cells, for example in a variety of tissue graftε including bone marrow tranεplantation. The existence of telomerase in stem or early progenitor cells does not preclude the utility of telomerase inhibition in cancer. Temporary telomerase inhibition in during cancer therapy in which tumor cells with short telomeres are induced to undergo crisiε (M2) εhould not have a εignificant biological effect on εtem cells since their telomereε are very long, and they divide very rarely in vivo .

Telomerase activity is useful as a marker of growth potential, particularly as to neoplastic cells, or progenitor cells, e.g.. embryonic stem cells. Human telomerase activity may be determined by measuring the rate of elongation of an appropriate repetitive sequence (primer) , having 2 or more, usually 3 or more, repeats of the telomere unit εequence, TTAGGG. The εequence iε labeled with a εpecific binding pair member at a convenient εite, e.g.. the 5'-terminuε, and the εpecific binding pair member allowε for separation of extended sequenceε. By uεing one or more radioactive nucleoεide triphoεphateε or other labeled nucleoside triphosphate, as described previously, one can measure the incorporated radioactivity aε cpm per unit weight of DNA as a function of unit of time, as a measure of telomerase activity. Any other detectable signal and label may also be used, e.g. , fluoreεcein.

The activity may be measured with cytoplasmic extractε, nuclear extracts, lysed cellε, whole cells, and the like. The particular sample which is employed and the manner of pretreatment will be primarily one of convenience. The pretreatment will be carried out under conditions which avoids denaturation of the telomeraεe, εo as to maintain the telomeraεe activity. The primer εequence will be εelected or labeled so as to allow it to be separated from any other DNA present in the sample. Thus, a haptenic label may be used to allow ready separation of the elongated sequence, which representε the telomeraεe activity of the εample. The nucleoside triphoεphateε which may be employed may include at leaεt one nucleoεide triphoεphate which iε labeled. The label will usually be radiolabel, but other labels may alεo be preεent. The labels may include specific binding pair members, where the reciprocal member may be labeled with fluorescers, enzymeε, or other detectable label. Alternatively, the nucleoεide triphoεphateε may be directly labeled with other labels, such aε fluoreεcent labels.

The sequence elongation usually will be carried out at a convenient temperature, generally from about 20°C to 40°C, and for a time sufficient to allow for at least about 100 bp to be added on the average to the initial sequence, generally about 30-90 minutes. After the incubation time to allow for the telomerase catalyzed elongation, the reaction may be terminated by any convenient meanε, εuch aε denaturation, e.g.. heating, addition of an inhibitor, rapid removal of the εequence by means of the label, and washing, or the like. The separated DNA may then be washed to remove any non-specific binding DNA, followed by a measurement of the label by any conventional meanε.

The determination of telomerase activity may be used in a wide variety of ways . It can be used to determine whether a cell is immortalized, e.g., when dealing with tiεεue associated with neoplasia. Thus, one can determine at the marginε of a tumor, whether the cellε have telomeraεe activity and may be immortalized. The preεence and activity of the telomerase may also be asεociated with εtaging of cancer or other diεeaseε. Other diagnostic interestε aεεociated with telomeraεe include meaεurement of activity aε an assay for efficacy in treatment regimens designated to inhibit the enzyme.

Other techniques for measuring telomerase activity can use antibodieε εpecific for the telomeraεe protein, where one may determine the amount of telomeraεe protein in a variety of wayε . For example, one may use polyclonal antisera bound to a surface of monoclonal antibody for a first epitope bound to a surface and labeled polyclonal antisera or labeled monoclonal antibody to a second epitope dispersed in a medium, where one can detect the amount of label bound to the surface as a result of the telomeraεe or εubunit thereof bridging between the two antibodies . Alternatively, one may provide for primers to the telomeraεe RNA and uεing reverεe tranεcriptaεe and the polymerase chain reaction, determine the preεence and amount of the telomerase RNA as indicative of the amount of telomerase present in the cells .

The following exampleε are offered by way of illuεtration and not by way of limitation. Examples The following are examples of specific aεpectε of the invention to merely illuεtrate thiε invention to those in the art. These examples are not limiting in the invention, but provide an indication of specific methodology useful in practice of the invention. They also provide clear indication of the utility of the invention and of the correlation between telomere length, telomerase activity and cellular senescence. Such correlation indicates to those in the art the breadth of the invention beyond these examples . Example 1: Telomere Length and Cell Proliferation

The effects of telomere length modulation on cellular proliferation were studied. An average of 50 bp are lost per cell division in somatic cells. The telomere end iε thought to have a εingle-εtranded region as follows (although the amount of overhang is unknown) :

5'TTAGGGTTAGGGTTAGGGTTAGGGTTAGGGTTAGGGTTAGGGTTAGGGTTAG GGTTA GGG 3'AATCCCAATCCC (Seq. ID No. 1)

Applicant postulated that loss of thiε single-stranded overhang should be significantly slowed if cellε were provided with a εynthetic oligonucleotide of the sequence CCCTAACCCTAA (Seq. ID No. 2) . This oligonucleotide εhould hybridize to the expoεed single- stranded region, and serve as a primer for DNA syntheεis by the normal DNA polymerase present in εomatic cellε. In thiε way, rather than shortening by an average of 50 bp per division, the telomeres may only shorten by a lesser amount per division, thus significantly extending the number of divisions required before telomere shortening induced cellular seneεcence. This hypotheεiε waε tested by measuring both the change in proliferative lifespan and rate of telomere εhortening in cultured cellε treated with thiε indicated oligonucleotide, versus control oligonucleotides.

The efficacy of the CTO-12 oligonucleotide (5' -CCCTAACCCTAA-3' Seq. ID No. 2) to reduce telomere shortening asεociated with cellular εeneεcence (Fig. 1) waε εtudied uεing target cellε cultured under εtandard cell culture conditionε in minimal esεential medium εupplemented with 10% fetal calf serum. The cellε were εubcultivated every four days by trypsinization upon reaching confluency and were fed new medium at subcultivation or every two days, whichever came first. Cells at various population doubling levels were seeded at 10,000 cellε per well and fed medium containing oligonucleotides at various concentrations. Oligonucleotideε studied were the cytidine-rich terminal oligonucleotide (CTO-12) , guanidine-rich terminal oligonucleotide-12 bp (GTO-12, having the sequence 5' -TTAGGGTTAGGG-3' (Seq. ID No. 3)), and a 12 baεe pair randomer with a random nucleotide in every poεition. Aε an additional control, cellε were fed identical medium without oligonucleotide. Cellε were fed oligonucleotide every 48 hourε from 10X εtockε. (Such oligonucleotideε may be modified to enhance εtability, e.g. , with phosphorothioates, dithioate and 2-0-methyl RNA.) In the case of phosphorothioates it would be desirable to use longer CTO primers such aε 5' -CCCTAACCCTAACCCT-3' , 5' -CCCTAACCCTAACCCTAA-3' , or 5' -CCCTAACCCTAACCCTAACC-3' .

Specifically, IMR-90 human lung fibroblaεtε with a proliferative capacity of approximately 55 population doubling (PD) were εeeded at PD45 at 10,000 cells per well in a 48 well tisεue culture diεh, and fed medium only or medium εupplemented with CTO-12 (at 1.0 μM and 0.1 μM) and 12 base pair randomer at 1.0 μM. As shown in Fig. 1, cells grown in medium without oligonucleotide, or with CTO-12 at less than 1.0 μM or with oligonucleotide of random sequence reached replicative senescence in a similar fashion at about 52 population doubling. Cells fed the CTO-12 oligonucleotide at 1.0 μM, however, continued to proliferate for approximately 10 doubling more than control cells. Example 2 : Inhibition of Telomerase in Cancer Cellε

One way by which cancer cells are able to escape cellular seneεcence is by regaining telomerase activity, which permits them to maintain the length of their telomeres in the face of multiple rounds of cell division. The enzyme telomerase contains an RNA complementary to TTAGGG, which allowε it to recognize the telomereε and extend them by the addition of additional TTAGGG repeatε. In fact, one aεεay for telomeraεe uses a TTAGGGTTAGGG primer and measureε the ability of cell extracts to syntheεiε a ladder of 6 bp additionε to this substrate. Telomerase activity in cancer cells is likely to be present in limiting amounts since telomere length iε relatively stable (thus only about 50 bp per telomere are added, so that lengthening and shortening are balanced) .

Applicant hypothesized that feeding cellε a synthetic TTAGGGTTAGGG oligonucleotide (Seq. ID No. 3) should competitively inhibit the ability of telomeraεe to elongate chromoεome endε, and thuε εhould lead to telomere shortening and senescence in cancer cells. Since somatic cells lack telomeraεe activity, the effects of this treatment should be strictly limited to cancer cells and the germ line.

Specifically, MDA 157 human breast cancer cellε with an immortal phenotype were εeeded at 10,000 cellε per well in 12 well tiεεue culture dishes and fed medium only or medium supplemented with GTO-12 (at 1.0 μM, 0.1 μM, and 0.01 μM) . As shown in Fig. 2, cells grown in medium without oligonucleotide, or with doses of lesε than 1.0 μM continued replicating in an immortal phenotype. Cellε fed the GTO-12 oligonucleotide, at 1.0 μM, however, ceased to proliferate after less than 10 doubling. Cells grown in the presence of 1.0 μM CTO-12 or 1.0 μM CTO-12 and 1.0 μM GTO-12 (G+C) continued to express the immortal phenotype suggesting that the GTO-12 oligonucleotide waε not intrinεically toxic (Fig. 3) . The lack of effect of the G+C mixture may reflect the CTO-12 oligonucleotide, competing with or baεe pairing with the GTO-12 oligonucleotide, thiε preventing itε inhibitory effect on the cancer cell telomeraεe. Example 3 : Telomere Length aε a Biomarker

In the U.S. and Weεtern Europe, atheroεclerosis iε the principal contributor to mortality from cardiovaεcular diεeaεeε (Ross, 314 N. Engl . J. Med. 488, 1986) . Atherosclerosis is characterized by the mural and focal formation of lipid and cell-rich leεionε or "plaqueε" on the intimal surfaces of arterial tissues. This iε followed by an age-dependent expanεion of the leεion into the lumen, potentially leading to occluεion and to myocardial and/or cerebral infarction (Hauεt, (1981) in Vaεcular Inj ury and Atheroεcleroεiε, ed. Moore, S. (Marcel Dekker Inc., New York), pp. 1-22; Ross and Glomset, 295(7) N. Engl. J. Med. 369, 1976; and Ross, 295(8) N_^ Engl. J. Med. 420, 1976) . Prominent among the mechanisms proposed to explain the pathogenesiε of atheroεcleroεiε iε the "response-to-injury" hypothesis (Roεε, 314 N. Engl. J. Med. 488, 1986; Moore, (1981) in Vaεcular Injury and Atheroεcleroεiε, ed. Moore, S. (Marcel Dekker Inc., New York), pp. 131-148; and Moore, 29(5) Lab. Invest. 478, 1971) in which repeated mechanical, hemodynamic and/or immunological injury to the endothelium is the initiating event.

A prediction of this hypothesis is that the intimal and medial tissue in the area comprising the atherosclerotic plaque will have a higher rate of cell turnover than the surrounding normal tisεue. Several lineε of evidence εupport thiε prediction. Roεs et al., (Ross and Glomset, 295(7) N. Engl. J. Med. 369, 1976; Ross, 295(8) N. Engl. J. Med. 420, 1976) showed that cultured εmooth muscle cells from fibrous plaques diεplayed lower responsiveness to growth serum when compared to cellε from the underlying medial layer. Moss and Benditt 78(2) (1973) Am. J. Pathol. 175, 1973, showed that the replicative life-span of cell cultureε from arterial plaques were equal to or leεε than the replicative life-εpans from cellε of nonplaque areas. Dartsch et al. , 10 Arteriosclerosiε 62, 1992, εhowed that human smooth muscle cells obtained from primary εtenoεing leεions became εenescent in culture far later than εmooth muεcle cells from restenosing lesions. These results suggest that cells derived from regions of atherosclerotic plaques undergo more cellular divisions than cells from non-plaque areas hence rendering them older and nearer to their maximum replicative capacity.

Thus, to understand the pathogenesis of atherosclerosiε, one must examine the alterationε in the behavior of cell turnover on and adjacent to the arterial leεionε. One requireε a biomarker for the cell turnover of intimal and medial tissue. Several workers have examined biomarkers for the progression of atherosclerosis or for the propensity of an individual to develop atherosclerosis. The former objective entailed the measurement of a number of biochemical compounds which are detected in the plasma but originate from the endothelium. Exampleε are serum Type III collagen (Bonnet et al., 18 Eur. J. Clin. Inveεt . 18, 1988) , von Willebrand' ε Factor (Factor VIII) (Baron et al., 10 Arterioεcleroεiε 1074, 1990) , cholesterol, triglycerides, apolipoprotein B (Stringer and Kakkar, 4 (1990) Eur. J. Vase. Surg. 513, 1990) , lipoprotein (a) (Breckenridge, 143 Can. Med. Assoc. J. 115, 1990; Mezdour et al. , 48 Ann. Biol. Clin. (Paris) 139, 1990; and Scanu, 14 Clin. Cardiol. 135, 1991) , endothelin (Lerman et al. , 325 N. Engl. J. Med. 997,

1991) and heparin-releaεable Platelet Factor 4

(Sadayaεu et al., 14 (1991) Clin. Cardiol. 725, 1991) .

A number of markerε originate from the cell εurface (Hanson et al. , 11 (1991) Arterioscler. Thro b. 745, 1991; and Cybulsky and Girnbrone, 251 Science 788, 1991) . Other markers monitor physiological aberrations as a result of atherogenesis (Vita et al. , 81 (1990) Circulation 491, 1990) . Candidate geneε used to delineate the RFLP profile of those suεceptible to atherogeneεiε (Sepehrnia et al. , 38 (1988) Hum. Hered. 136, 1988; and Chamberlain and Galton, 46 Br. Med. Bull . 917, 1990) have alεo been established. However, there have been relatively few markers developed to monitor directly cell turnover.

Applicant now εhows that telomere length may serve as a biomarker of cell turnover in tissues involved in atherogenesiε . The reεultε εhow that endothelial cellε loεe telomereε in vi tro aε a function of replicative age and that in vivo telomere loεε iε generally greater for tiεεueε of the atheroεclerotic plaqueε compared to control tiεεue from non-plaque regionε .

In general, telomere lengthε were assessed by Southern analysiε of terminal reεtriction fragments (TRF, generated through Hinfl/Rεal digestion of human genomic DNA. TRFs were resolved by gel electrophoresiε and hybridized with a telomeric oligonucleotide (³²P- (CCCTAA)₃) (Seq. ID No. 4) . Mean TRF length decreaεed as a function of population doubling in human endothelial cell cultureε from umbilical veins (m=-190 bp/PD, P=0.01), and as a function of donor age in iliac arteries (m=-120 bp/PD, P=0.05) and iliac veins (m=-160 bp/PD, P=0.05) . Thus, mean TRF length decreased with the in vitro age of all cell cultures. When early passage cell cultureε were aεsessed for mean TRF length as a function of donor age, there was a significant decreaεe for iliac arterieε (m=-102 bp/y, P=0.01) but not for iliac vein (m=47 bp/y, P=0.14) . Mean TRF length of medial tissue decreased significantly

(P=0.05) aε a function of donor age. Intimal tiεsues from one individual who diεplayed extensive development of atherosclerotic plaques possessed mean TRF lengths close to those observed for senescent cells in vitro (-6 kbp) . Theεe obεervationε indicate that telomere εize indeed serves as a biomarker for the replicative history of intima and media and that replicative senescence of endothelial cells is involved in atherogenesis.

Specifically, the following materialε and methods were used to achieve the results noted below. Endothelial Cell Cultures

Human umbilical vein endothelial cellε (HUVEC) were obtained from Dr. Thomaε Maciag of the Jerome H. Holland Laboratory of the American Red Croεε. Human endothelial cells from the iliac arteries and iliac veins were obtained from the Cell Repository of the National Institute of Aging (Camden, New Jersey) . Cells were grown at 37°C in 5% C0₂ on 100 mm tissue plates whose interiors were treated with an overnight coating of 0.4% gelatin (37°C) . The supplemented media consiεted of M199, 15% fetal bovine serum, 5 U/ml heparin and 20 μg/ml crude Endothelial Cell Growth Supplement (Collaborative Reεearch) or crude Endothelial Cell Growth Factor (Boehringer-Mannheim) . Cultureε were trypsinized (0.05%, 3 minutes) at confluence, reseeded at 25% of the final cell density and refed every 2-3 days. Tissue Samples

Tissue samples from the aortic arch, abdominal aorta, iliac artery and iliac vein were obtained from autopsies at the Department of Pathology, Health Sciences Center, McMaεter Univerεity. Poεt-mortem times ranged from 5 to 8 hours. The intima was obtained by cutting open the arteries or veins and carefully scraping off the lumenal surface with a No. 10 scalpel (Lance Blades, Sheffield) (Ryan, 56 Envir. Health Per. 103, 1984) . The resulting material was either treated directly for extraction of DNA or proceεεed for cell culture.

The adventitial layer was removed by cutting or scraping the non-lumenal side of the veεεel. The remaining medial layer was prepared for DNA extraction by freezing it in liquid-N₂ and grinding it in a liquid- N₂ chilled mortar and pestle (Kennedy et al. , 158 Exp. Cell Res. 445, 1985) . After the tisεue was ground to a powder, 5 ml of frozen digestion Buffer (10 mM Tris; 100 mM NaCl; 25 mM EDTA; 0.5% SDS; pH 8.0) waε added and ground into the powderized tiεεue. The powder waε then tranεferred to a 50 ml Falcon tube and incubated at 48°C until thawed. Proteinaεe K (10 mg/ml) was added to a final concentration of 0.2 mg/ml. After a 12-16 hour incubation, the solution was removed from the water bath and either prepared for DNA extraction or stored at 20°C.

Extraction and Restriction Enzyme Digestion of Genomic DNA

DNA waε extracted as described previously

(Harley et al., 345 Nature 458, 1990; Allsopp et al . ,

89 Proc. Natl. Acad. Sci. USA 10114, 1992) . In brief, proteinase K-digeεted lyεateε were extracted twice with one volume of phenol:chloroform:iεoamyl alcohol

(25:24:1) and once with chloroform. Nucleic acid waε precipitated by adding 2 volumes of 100% EtOH to the aqueous layer, washed once with 70% EtOH and finally reεuεpended in 100-200 μl of 10 mM Tris-HCl, 1 mM EDTA, pH 7.5. DNA was quantified by fluorometry and 1 μg was digeεted with 1 unit each of Hinfl /Rεal for 3-24 hourε at 37oC. Complete digeεtion was monitored by gel electrophoresis. The integrity of the DNA before and after digestion was monitored in control experimentε by gel electrophoreεis .

Southern Blot Hybridization

Electrophoresiε of digeεted genomic DNA was performed in 0.5% agarose gelε in a εtandard Triε, εodium borate, EDTA buffer for a total of 650-700 V/hr aε described previously (Harley et al. , 345 Nature 458, 1990; Allsopp et al. , 89 Proc. Natl. Acad. Sci. USA 10114, 1992) . After electrophoresiε, the gel waε placed onto 3 mm Whatman filter paper and dried under vacuum for 25 minuteε at 60^*C. Gelε were denatured by εoaking in 0.5 M NaOH, 1.5 M NaCl for 10 minuteε at room temperature and then neutralized through immerεion in 0.5 M Triε, 1.5 M NaCl. Genomic DNA was immersed in εtandard hybridization εolution (Harley et al. , 345 Nature 458, 1990) (6X SSC) with the telomeric ³²P- (CCCTAA)₃ probe (Seq. ID No. 4) for 12-16 hourε at 37^*C. The telomeric εmearε were visualized through autoradiography on pre-flashed (0D₅₄₅ = 0.15) Kodak XAR-5 film. The mean lengths of the terminal restriction fragments (TRFε) were calculated from denεitometric εcans of the developed films as described previously (Harley et al . , 345 Nature 458, 1990) . In vi tro Results

To determine the feasibility of employing telomere length as a biomarker for cell turnover in atherosclerosis, we first examined the change in telomere length in cultured endothelial cells where cell diviεion can be directly monitored in vitro . The DNA waε digested with Hinfl and Rεal , and the resulting terminal restriction fragments (TRF) were subjected to Southern analysis. As in human skin fibroblaεts (Allsopp et al., 89 Proc. Natl. Acad. Sci. USA 10114, 1992) , mean TRF length decreased as a function of population doubling (PD) . Thus, telomere length decreaεeε with in vi tro age of human umbilical vein endothelial cells. Mean TRF length decreased linearly (P=0.01) at a rate of 190 ± 10 bp/PD (εee Fig. 4) . The Y-intercept, which εignifieε the mean TRF at 0 PDL iε 14.0 kbp while mean TRF at senescence waε 5.7 ± 0.4 kbp.

To prove that telomere length decrease occurred in endothelial cells from other arterial and venous sources, mean TRF length versus population doubling level (PDL) was determined for several strains of endothelial cellε from human iliac artery and human iliac vein. In both iliac arteries and iliac veins there was a significant (P=0.05) linear decrease in mean TRF length with age of culture: 120 ± 60 bp per population doubling for the iliac artery and 160 ± 30 bp per population doubling for the iliac veins from endothelial cells. In vivo Reεultε

Formation of atheroεclerotic plaqueε occurs more often in the iliac artery than in the iliac vein

(Crawford, (1982) Pathology of Atheroεclerosiε

(Butterworth and Co. Ltd., U.K.), p. 187-199), thus it is expected that turnover of intimal tiεsue in vivo from the iliac artery should be greater than that from the iliac veins. To test this, nine different strainε of endothelial cell cultureε from iliac arterieε and veins of donors ranging in age from 14-58 yearε of age were cultivated and TRF lengthε from the earliest possible PDL were determined (Fig. 5) . Consistent with the hypothesis of greater cell turnover in vivo in arteries than in veins, the rate of decrease in mean TRF length, was significant over the age range 20-60 yearε for iliac arteries (-100 bp/yr, P=0.01) and greater than for the iliac veins (-47 bp/yr, P=0.14). Among the nine strains of endothelial cells, there were cultures from the iliac artery and iliac vein from the same individuals for 3 of the donors, aged 21, 47 and 49 years. There was a significantly shorter mean TRF length in the cultures of iliac artery cells as compared to the venous cells for the two older donors. The younger donor showed no significant difference in mean TRF length between the two cultures, possibly reflecting relatively little difference in cell turnover between the vesεelε of the 21-year old donor.

Differenceε in mean TRF length of the cell cultureε from iliac arterieε and iliac veins in donors of different ages will reflect not only differences in original mean TRF length of the primary tiεεues but also differences in the rate of telomere loεs between the different cultures in vitro during the time required to collect sufficient cells for analysiε (approximately 5-10 PDL) . To determine if there iε a relationship between cell turnover and the extent of atherosclerotic plaque formation, we examined mean TRF length in primary tissue. Autopsieε from 3, 11, 12, 14, 18, 26, 75-year old femaleε and a 77-year old male were performed. Sectionε of the aortic arch, abdominal aorta, iliac artery and iliac vein were taken and the intimal and medial tiεsues εeparated and assessed for TRF length.

Sufficient intimal tissue could be obtained from the aortic arch, abdominal aorta, iliac arteries and iliac veins of 3 donors (aged 27, 75 and 77 years) for TRF analysis . There was a εtriking difference between the mean TRF lengthε averaged over these εiteε in the 27-year old female (10.4 ± 0.7 kbp) verεuε the 75-year old (8.8 + 0.6 kbp) and the 77-year old male (6.3 + 0.4kbp) . It is noteworthy that the 77-year old male had extensive atherosclerotic lesions in his vasculature and that the mean TRF length of hiε intimal tissue is close to that of endothelial cells, at senescence in vitro (approximately 6 kbp, Fig. 4) .

Fig. 6 shows that mean TRF of medial tissue (from the aortic arch) decreaseε with donor age at a εmall but εignificant rate (47 bp/yr, P=0.05) . Thus, medial cells turnover in vivo occurs at a rate less than that of the venous or arterial endothelial cells .

In general, telomere losε in medial tissue underlying an atherosclerotic plaque was greater than those in non-plaque regions (Table 1) . With the 75- year old female, mean TRF was significantly reduced in medial DNA from the plaque regions verεuε the non- plaque regionε of both the aortic arch (P=0.04) and the abdominal aorta (P=0.01) . For the 77-year old male, this was observed in the abdominal aorta (P=0.01) . TABLE 1

Mean TRF values for primary medial tissues of plaque and non-plaque areas

Plaque Region Non-Plaque Region P 75-year old Donor Aortic Arch 10.2+0.5 11.1 ±0.1 0.04

Abdominal Aorta 9.5 ± 0.6 11. O ± O.1 0.01

77-year old Donor

Aortic Arch 8.2 ± 0.4 8.4 ± 0.2 NS

Abdominal Aorta 7.1 ± 0.1 8.2 ± 0.4 0.01 Theεe results show that mean TRF length decreases as a function of donor age for primary medial and intimal tissue, suggesting that cell turnover does occur in cardiovascular tiεεue. The decreaεe in mean TRF length for plaque regionε verεuε clear regionε of medial tiεεue from the εame blood veεεel iε conεiεtent with augmented cell turnover of tissue asεociated with atheroεclerotic plaqueε . Thuε, the results indicate that measurement of telomere length provides a biomarker for alterations of cellular turnover in tisεues associated with cardiovascular diseaseε, i.e. , cells of the intima and media.

Measurement of telomere length is a direct register of proliferative history but to obtain telomeric DNA one must obtain a biopsy of endothelial tiεεue. Since removal of the endothelium in itself can induce plaque formation, the biopsy strategy obviously entails ethical and practical problems. Based upon experience with autopsy samples one requires a minimal area of 1 cm² in order to perform a Southern analyεiε aε deεcribed in thiε paper. For a practical biopsy, this is untenable. A detection technique to circumvent thiε problem may be confocal fluoreεcent microscopy. Example 4 : Simplified Test for Telomere Length Telomere length haε been found to be the best predictor of the remaining lifespan of cells cultured from donors of different ageε. The ability to meaεure telomere length thus has significant clinical use. Because of their simple repetitive nature, telomeres lack DNA sequences recognized by many restriction enzymes . One way to measure telomere length is to digest DNA with reεtriction enzymes with 4-base recognition siteε, which cutε moεt of the DNA into very εmall pieceε and leaveε the telomeres in relative large TRFs (Terminal Restriction Fragments) . A Southern blot of the DNA is then probed with a radioactive TTAGGGTTAGGGTTAGGG (Seq. ID No. 5) oligonucleotide, and the size of the TRF determined.

A much simpler method to measure telomere length exploits the fact that the telomere sequence lacks guanidine residues in the C-rich strand. Genomic DNA can be melted and mixed with the DNA synthesiε primer CCCTAACCCTAACCCTAACCCTAA (Seq. ID No. 6) in the preεence of DNA polymeraεe and only three deoxynucleotides (dATP, dTTP and radioactive dCTP) . Rare complementary sequenceε εcattered throughout the genome would fail to extend due to the lack of dGTP. The length of the extended DNA can then be determined from a simple gel electrophoresis. The amount of DNA synthesized (counts incorporated per μg of DNA) will be directly proportional to telomere length, and for diagnoεtic purposeε a εimple measure of radioactivity would then suffice to quantitate telomere length. Example 5: Identification of DNA Sequences Near Telomeres

There are good reasons to believe that the regulatory factors that control cellular and organismal senescence are located near telomeres, and are themεelves regulated by the length of the adjacent telomere. It is thuε important to identify and clone them in order to be able to understand and manipulate the aging proceεε. In addition, there iε great interest in identifying unique telomeric DNA within the human genome project, since telomeric markers for mapping purposes are lacking for the ends of the chromosomes. In one method, large telomeric DNA is purified aε follows. A biotinylated CCCTAACCCTAA (Seq. ID No. 7) oligonucleotide iε used to prime DNA syntheεiε in double-εtranded genomic DNA. The only εequenceε with which this oligonucleotide can anneal will be the single-stranded base overhangs at telomere endε. The extended DNA will then be digeεted with a reεtriction enzyme such as No tl to produce large restriction fragmentε. Biotinylated fragments are retrieved using streptavidin coated magnetic beads, and analyzed by pulsed field electrophoresis. 46 fragments (one for each end of the 23 human chromosomeε) are produced. Multiple strategies can be uεed to purεue the εucceεεful iεolation of large telomeric DNA. The DNA can be labeled and uεed to screen cDNA libraries in order to identify genes located near telomeres. The expression of these cDNAs can then be examined in young versuε old cells in order to identify those which are differentially expressed as a function of cellular seneεcence, and which are thuε candidateε to be regulatory factorε that control aging.

The purified telomeric DNA can alεo be digested with additional restriction enzymes, mixed with 100-fold excess of genomic DNA, melted and reannealed. Under these circumεtanceε, the repetitive εequences in the telomeric DNA will anneal with genomic DNA while unique sequenceε in the purified DNA will εelf-anneal. Only the self-annealed unique sequences will contain restriction overhangs at each end, and thuε a εimple cloning of the annealed DNA will reεult in the εucceεεful cloning of only unique fragmentε. Example 6: Telomere Loss in Down's Syndrome Patients Losε of telomeric DNA from human chromoεomes may ultimately cause cell cycle exit during replicative seneεcence. Since lymphocyteε have a limited replicative capacity and blood cells were previously shown to lose telomeric DNA during aging in vivo, we wished to determine whether accelerated telomere loss is associated with the premature immunosenescence of lymphocytes in individuals with Down's Syndrome (DS) , and whether telomeric DNA iε alεo loεt during aging of lymphocytes in vi tro . To investigate the effectε of aging and triεomy 21 on telomere loεε in vivo, genomic DNA was isolated from peripheral blood lymphocytes of 140 individualε (0-107 y) and 21 DS patientε (0-45 y) . Digeεtion with reεtriction enzymeε Hinfl and Rεal generated terminal restriction fragments (TRFs) which can be detected by Southern analysiε uεing a telomere- specific probe, (³P- (CCCTAA)₃) . The rate of telomere loss was calculated from the decrease in mean TRF length aε a function of donor age. DS patientε showed a significantly higher rate of telomere loεε with donor age (133 ± 15 bp/y) compared to age-matched controls (41 ± 7.7 bp/y) (P<0.0005), indicating that accelerated telomere loss is a biomarker of premature immunosenescence of DS patients, and may play a role in this process.

Telomere loss during aging in vi tro was calculated for lymphocytes from two normal individuals grown in culture for 20-30 population doubling. The rate of telomere loss was 90 bp/cell doubling, that is, it was comparable to that seen in other somatic cells. Telomere lengths of lymphocytes from centenarians and from older DS patients were similar to those of senescent lymphocytes in culture, which suggests that replicative senescence could partially account for aging of the immune syεtem in DS patientε and elderly individuals. The following materials and methods were used to obtain the results provided below.

Culture of Human Peripheral Blood T Lymphocytes

Adult peripheral blood εampleε were collected, and mononuclear cellε were isolated by Ficoll-Hypaque gradient centrifugation then cryopreserved in liquid nitrogen. Cultures were initiated by mixing IO^"6 mononuclear cells with IO⁶ irradiated (8000 Rad) lymphoblastoid cells (Epstein- Barr viruε tranεformed B cellε) , or 10^s mononuclear cellε with 10 μg/ml phytohemagglutinin (PHA-P, Difco) in each well of a 48-well cluster plate (Costar) . After 8 to 11 days, cells were washed and plated in 2 ml wells of 24-well cluster plateε at a concentration of 2-4 x 10⁵/ml. Cultureε were paεεaged every three to four dayε, or whenever viable cell concentration (determined by trypan blue excluεion) reached ≥ 8xl0⁵/ml. Cultureε were terminated when they showed no proliferative response to irradiated lymphoblastoid cells and/or when there were no viable cells present in the entire visual field of the haemocytometer. Once tranεferred to the 2 ml wellε, cells were continuously exposed to 25 U/ml of recombinant interleukin-2 (Amgen) . The media used were (a) RPMI (Irvine Scientific) supplemented with 10 to 20% fetal calf serum, 2 mM glutamine, and 1 mM Hepes,- (b) AIM V™, a DMEM/nutrient mixture F-12 basal medium, containing purified human albumin, transferrin, and recombinant insulin (Gibco) , supplemented with 25% Ex-cyte (an aqueous mixture of lipoprotein, cholesterol, phospholipids, and fatty acids, (Miles Diagnostics) . At each cell passage, the number of population doubling (PD) was calculated according to the formula: PD = In (final viable cell no. initial cell no.)/ln2. Isolation of DNA PBLs (including = 15% monocytes) were isolated using Ficoll-Hypaque gradient centrifugation (Boyum et al., 21(97) Scan. J. Clin. Lab. Invest. 77, 1968) and waεhed 3 timeε in PBS . Cell pelletε were resuspended in 500 μl of proteinaεe K digestion buffer (100 mM NaCl, 10 mM Triε pH 8, 5 mM EDTA, 0.5% SDS) containing 0.1 mg/ml proteinaεe K and incubated at 48"C overnight. Lyεateε were extracted twice with phenol/chloroformiεoamyl alcohol (25:24:1 v/v/v) and once with chloroform. DNA waε precipitated with 95% ethanol and dissolved in TE (10 mM Triε, 1 mM EDTA, ρH=8) .

Analyεis of Telomeric DNA

Genomic DNA (10 μg) waε digeεted with Hinfl and . sal (BRL) (20 U each), re-extracted as above, precipitated with 95% ethanol, washed with 70% ethanol, disεolved in 50 μl TE, and quantified by fluorometry. One μg of digeεted DNA was resolved by electrophoresis in 0.5% (w/v) agarose gelε poured on Gel Bound (FMC Bioproducts) for 700 V-h. Gels were dried at 60'C for 30 minutes, denatured, neutralized, and probed with 5' end-labeled ³P- (CCCTAA) as deεcribed above. Autoradiograms exposed within the linear range of signal response were scanned with a Hoefer densitometer. The signal was digitized and subdivided into 1 kbp intervals from 2 kbp to 21 kbp for calculation of the mean TRF length (L) using the formula L=Σ, (ODi- i) /ΣOΩ_{i t} where ODi = integrated signal in interval i, and L=TRF length at the mid-point of interval i.

TRF Length vs. Age

When measured as a function of donor age, mean TRF length in PBS of 140 unrelated normal individuals (aged 0-107 y) declined at a rate of 41 ± 2.6 bp/y (p<0.00005, r=0.83) . This rate of TRF loss for PBLs iε cloεe to that previouεly found for peripheral blood cells by Hastie et al. , 346 Nature 866, 1990. When our data were separated according to gender it was noticed that males loεt telomeric DNA at a rate εlightly faεter than that of females (50 ± 4.2 vs 40 ± 3.6 bp/y), but this difference did not reach statiεtical significance

(p=0.1) . The 18 centenarians (aged 99-107 y) among our population of normal individuals had a mean TRF length of 5.28 + 0.4 kbp (Fig. 7) . Intereεtingly, the standard deviation of mean TRF values for the centenarians (0.4 kbp) was much smaller than that of other age groups. Although it iε poεεible that thiε represents selection of a more homogeneous population of cells with age, it is also possible that the group of centenarians were lesε genetically diverεe than the younger populationε in our εtudy. Mean TRF length waε alεo analyzed in PBLε of 21 Down's Syndrome individuals (aged 2-45 y) and the rate of loss was compared to 68 age-matched controls (aged 0-43 y) . We found that cellε from DS patientε showed a significantly greater rate of telomere losε (133 ± 15 bp/y vs 41 ± 7.7 bp/y; one tailed t-test, t=5.71, p<0.0005) (Fig. 8).

To determine the rate of telomere loεs as a function of cell doubling, we cultured normal lymphocytes from 2 individuals in vi tro until replicative senescence and measured mean TRF length at εeveral population doubling levels (Fig. 9) . Mean TRF length decreased 90 bp/population doubling in these strainε, within the range obεerved for other human εomatic cell typeε. The mean TRF length at senescence for the lymphocyte cell εtrainε shown here and one other analyzed at terminal pasεage (Fig. 9), was 5.1 ± 0.35 kbp. The observed TRF values in vivo for PBLs of centenarianε (5.3 ± 0.4 kbp) and old DS patientε (4.89 ± 0.59 kbp), were cloεe to thiε value, εuggesting that a fraction of the cells from these individuals were close to the limit of their replicative capacity.

The results showing that telomereε in PBLε from normal individualε shorten during aging in vivo and in vitro extend similar observations on human fibroblasts (Harley et al. , 345 Nature 458, 1990) and support the hypotheεiε that telomere loεs is involved in replicative senescence. We also found that in Down's Syndrome, the rate of telomere losε in PBS in vivo was significantly higher than that in age-matched normal donors. Thuε, accelerated telomere loεε in PBS of trisomy 21, a syndrome characterized by premature immunosenescence and other features of accelerated aging (Martin, "Genetic Syndromes in Man with Potential Relevance to the Pathobiology of Aging", in: Genetic Effects on Aging, Bergsma, D. and Harrison D.E. (eds.) , pp. 5-39, Birth Defects: Original article serieε, no. 14, New York: Alan R. Liεε (1978)), could reflect early εeneεcence of lymphocytes.

The increased rate of telomere loss in PBS from DS patients could reflect a higher turnover rate of cells in vivo due to reduced viability of the trisomy 21 cells. However, it is alεo poεεible that the rate of telomere loss in PBS from DS patients is greater per cell doubling than that in normal individuals. The pathology of DS is similar in many ways to normal aging. Premature senescence of the immune system posεibly playε a role in thiε εimilarity since DS patients have a high incidence of cancer and suffer from autoimmunit . In support of this idea, lymphocytes of older DS patients and old individuals share several characteristics, including diminished response of T- cells to activate and proliferate in response to antigen, low replicative capacity, and reduced B- and T-cell counts (Franceschi et al., 621 Ann. NY Acad. Sci. 428, 1991) . Our finding that telomere length decreased faster in DS patientε than normal individualε, and that the mean TRF length in centenarianε and old DS patients in vivo were similar to that of seneεcent lymphocytes in vi tro (=5 kbp) 1 extends these observations. Moreover, these data suggest that replicative senescence within the lymphoid lineage in vivo contributes to the compromised immune system of both elderly individuals and Down's Syndrome patients. Example 7: Ovarian Cancer and Telomerase Activity

The following is an example of a method by which telomerase activity is εhown to correlate with the presence of cancer cells. In addition, the length of TRF waε determined as an indication of the presence of tumor cells. Generally, it was found that tumor cells had significantly lower TRF values than surrounding normal cells, and had telomerase activity. Thus, these two features are markers for the presence of tumor cells.

The following methods were used to obtain theεe reεultε: Separation of Tumor and Non-tumor Cellε

In one method, aεcitic fluid waε obtained by either diagnoεtic laparotomy or therapeutic paracenteεiε (from patientε diagnoεed aε having ovarian carcinoma) , and centrifuged at 600 xg for 10 minuteε at 4°C. The cell pellet waε waεhed twice in 10 to 30 ml of phosphate buffered saline (PBS: 2.7 mM KC1, 1.5 mM KH₂P0₄, 137 mM NaCl and 8 mM Na₂HP0₄) and centrifuged at 570 xg for 4 minutes at 4^*C. After the final wash the cell pellet was resuspended in 20 ml of PBS and filtered through a 30 or 10 μm nylon mesh filter

(Spectrum) which retains the tumor clumps but not εingle cellε. The filterε were backwaεhed to liberate highly purified tumor clumpε. The flow-through waε a combination of fibroblasts, lymphocytes and tumor cells.

In another method aεcitic fluid cellε were collected and waεhed as deεcribed above. The cellular pellet waε resuspended in a-MEM with 10% fetal calf serum and cultured in 150 mm dishes. After 12 hours the media was removed and new plates were used to εeparate the adhering fibroblasts from the non-adhering cells in the medium. After 12 hours the media containing moεtly tumor clumpε waε removed from the εecond plates and allowed to adhere in DMA F12 medium supplemented with 3% fetal calf serum, 5 ng/ml EGF, 5 μg/ml insulin, 10 μg/ml human transferrin, 5xl0^"s M phoεphoethanolamine and 5xl0^"s M ethanolamine. These tumor cellε were cultured for DNA analyεis and S100 extracts . DNA Extraction

Cellε were lysed and proteins were digeεted in 10 mM Triε-HCl (pH 8.0) , 100 mM NaCl, 25 mM EDTA, 0.5% SDS, 0.1 mg/ml proteinaεe K at 48°C overnight. Following 2 extractionε with phenol and 1 with chloroform, DNA waε precipitated with ethanol and diεsolved in 10 mM Tris-HCl (pH 8.0) , 1 mM EDTA (TE) . Determination of TRF Length and Amount of

Telomeric DNA

Genomic DNA waε digeεted with Hinfl and Rεal , extracted and precipitated aε above, and redissolved in TE. DNA concentration was meaεured by fluorometry (Morgan et al . , 7 Nucleic Acids Res. 547, 1979) . DNA samples (1 μg each) were loaded onto a 0.5% agarose gel and electrophoresed for 13 hours at 90 V. The gel was dried at 60°C for 30 minutes, denatured in 1.5 M NaCl and 0.5 M NaOH for 15 minutes, neutralized in 1.5 M NaCl, 0.5 M Tris-HCl (pH 8.0) for 10 minutes and hybridized to a 5' ³²P (CCCTAA)₃ telomeric probe in 5x SSC

(750 mM NaCl and 75 mM sodium citrate) , 5x Denhart's εolution (Maniatiε et al. , Molecular Cloning: A

Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor (1982)) and 0.Ix P wash (0.5 mM pyrophoεphate, 10 mM Na₂HP0₄) at 37°C for 12 hourε. Following three high εtringency waεhes in 0.24x SSC at 20-22°C (7 minuteε each) , the gel waε autoradiographed on pre-flaεhed (OD = 0.15) Kodak XAR-5 X-ray films for 3 days with enhancing εcreenε . Each lane was εcanned with a denεitometer and the data uεed to determine the amount of telomeric DNA and the mean TRF length aε previouεly deεcribed (Harley et al . , 345 Nature 458, 1990) . Preparationε of S-100 Cell Extracts

A minimum of 6xl0⁸ cellε were used for each extract. Ascitic fluid or purified ascitic fluid tumor cells (by the first method described above) were centrifuged at 570 xg for 4 minutes at 4°C. Ascitic fluid tumor cells separated by the second method described above (grown in monolayer) were harvested by scraping with a rubber policeman, and centrifuged aε above. The pelletε were rinsed twice in cold PBS followed by centrifugation as above. The final pellet was rinsed in cold 2.3x Hypo buffer (lx Hypo buffer: 10 mM Hepeε (pH 8.0)) , 3 mM KC1, 1 mM MgCl₂, 1 mM DTT, 0.1 mM PMSF and 10 U/ml of RNAsin, 1 μM leupeptin and 10 μM pepstatin A, centrifuged for 5 minutes and resuspended in 0.75 volumes of 2.3x Hypo buffer. After incubation on ice for 10 minutes the sample was tranεferred to an ice cold 7 or 1 ml Dounce homogenizer and homogenized on ice using a B pestle (25-55 μm clearance) . After a further 30 minutes on ice the samples having a volume larger than 1 ml were centrifuged for 10 minutes at 10,000 rpm (16,000xg) at 4°C in a Beckman J3-13.1 swinging bucket rotor. One- fiftieth volume of 5 M NaCl waε added, and the samples supernatant were centrifuged for 1 hour at 38,000 rpm

(100,000xg) at 4°C in a Beckman Ti50 rotor. Glycerol was added to a final concentration of 20% and the extract aliquoted and stored at -70°C. Samples lesε than 1 ml were centrifuged at 55,000 rpm for 1 hour at 4°C in a TLA 100.2 rotor (Beckman) and NaCl and glycerol were added to the εupernatant aε above. Protein concentration in a typical extract waε approximately 4 mg/ml. Telomerase Assay

Telomerase activity waε assayed by a modification of the method of Morin, 59 Cell 521, 1989. Aliquots (20 μl) of S-100 cell extract were diluted to a final volume of 40 μl containing 2 mM dATP, 2 mM dTTP, 1 mM MgCl₂, 1 μM (TTAGGG) ₃ primer, 3.13 μM (50 μCi) a-³²P-dGTP (400 Ci/mmole) , 1 mM spermidine, 5 mM β- mercaptoethanol, 50 mM potaεεium acetate, and 50 mM Tris-acetate (pH 8.5) . In εome experimentε reaction volumeε were doubled. The reactionε were incubated for 60 minuteε at 30°C and stopped by addition of 50 μl of 20 mM EDTA and 10 mM Tris-HCl (pH 7.5) containing 0.1 mg/ml RNAseA, followed by incubation for 15 minuteε at 37°C. To eliminate proteins, 50 μl of 0.3 mg/ml Proteinase K in 10 mM Tris-HCl (pH 7.5) , 0.5% SDS waε added for 10 minutes at 37°C. Following extraction with phenol and chloroform, unincorporated a-³²P-dGTP was separated by centrifuging the samples for 4 minutes at 500 g in a swinging bucket rotor through NICK SPIN columns (Pharmacia) . DNA waε precipitated by the addition of 5.3 μl of 4 M NaCl, 4 μg of carrier tRNA and 500 μl of ethanol at -20°C. DNA pellets were resuspended in 3 μl of formamide loading dye, boiled for 1 minute, chilled on ice and loaded onto an 8% polyacrylamide, 7 M urea sequencing gel and run at 1700 V for 2 hours using 0.6X TBE buffer. Dried gelε were expoεed to Kodak XAR-5 pre-flaεhed film at -70°C with enhancing εcreen or to phoεphoimager screens (Molecular Dynamics) for 7 days.

The results of the above experiments are shown in tables 2 and 3 below:

Table 2 : Characteristicε of ATCC Ovarian Carcinoma Cell Lines

Cell line Mean TRF Length (kbp) Telomeraεe Activity

HEY stable at 3.7 +

CAOV-3 stable at 3.7 N.D.

SKOV-3 Increaεes at 60 bp/pd N.D.

Table 3 : Characteristics of Ovarian Carcinoma Tumor Cells from Ascitic Fluid

Patient Description Mean TRF Telomerase Length (kbp) Activity

Pres-3 Purified tumor cells 3.7 + Mac-2 Purified tumor cells 3.7 N.D.

Sib-1 Purified tumor cells 4.2 N.D.

Ric 207 Purified tumor cells 3.3 N.D.

Cra-1 Purified tumor cells 5.2 N.D.

Ing-1 Purified tumor cells 5.8 N.D.

Lep-1 Purified tumor cellε 5.8 N.D.

Lep-4 Purified tumor cells 5.6 N.D.

Sol-1 Purified tumor cells 5.6 N.D.

Rud-1 Ascitic : Eluid < Dells 3.4 +

Murr-1 Ascitic : Eluid ( Dells 3.8 +

Dem-1 Ascitic : Eluid < Dells N.D. +

Cas-1 Ascitic : Eluid < Dells 5.3 +

Wad-1,2 Ascitic Eluid < Dells 4.9 N.D.^*

N.D. = not determined

* High background precluded detection

Table 4 shows the TRF length of cells from ascitic fluid. A minimum of 2 autoradiographs were εcanned with a denεitometer over the εize range 2 - 21 kbp, and the denεitometric valueε uεed to determine mean TRF length in kbp. Average εtandard deviation of the data waε 0.5 kbp with the largest deviation being 2 kbp. The value following the three character patient code refers to the paracenteεiε number (i.e. , OCl-1 iε the firεt εample from patient OC1) . Sampleε defined aε E (early) were obtained near the time of preεentation while εampleε L (late) were obtained near death. Paracenteεeε were performed 4 to 15 timeε over the courεe of 4 to 22 monthε. Table 4

Table 5 shows the telomerase activity in normal and tumor cellε. Leukocyteε and acsites cells were isolated and aεcitic fluid cells fractionated into normal and tumor fractions and asεayed for telomerase activity. Protein concntration in all extracts waε

<2 mg/ml, i.e., 20 fold higher than the loweεt concentration at which activity waε detected in control 293 CSH extract. Table 5

In the TRF aεεay, each tumor clump had significantly lower TRF lengths than asεociated normal cellε. (See Fig. 10) .

Referring to Fig. 41, data iε compiled εhowing the reεultε of telomeraεe assays of normal cells and tissues and cancer cell lines and tissueε. As can be seen in the figure, normal somatic cellε generally lack telomeraεe activity, with the exception of hematopoietic εtem cellε. Normal germ-line cellε such as mouse embryonic stem cellε alεo εhowed telomeraεe activity. In contraεt to normal cells, immortalized cancer cell lines display telomerase activity as does various εampleε of tumor tiεεue.

In the telomerase asεay, significantly greater telomerase activity waε evident in the aεcitic fluid of certain patientε than in the control tumor lines HEY and PRES, or the control cell line 293 CSH (Fig. 11, 33) . Example 8 : Effect of HIV Infection on TRF Length

HIV infection leads to an acute viral infection manifesting itself aε a viruε-like εyndrome, followed by a prolonged period of latency characterized by an abεence of εignε and symptoms. During this prolonged asymptomatic period (lasting usually 7-10 years) , there is no diagnostic available for staging the course of the infection other than the presence or absence of antibodies to viral coat proteinε . This does little to stage the diεease or to help the physician measure the effectiveness of prophylactic agents .

While Meyaard et al. , 257 Science 217, 1992, propose a programmed cell death for CD4⁺ and CD8⁺ cellε of an HIV-infected individual, we propoεe that during those 7 to 10 years the immune system iε able to keep the infection relatively repreεεed, but there iε markedly increaεed turnover of the infected CD4⁺ T- cells . This may be due in part to viral-mediated cell destruction. We propose that this essentially accelerates the replicative εeneεcence of this particular subpopulation of T-cellε, and with time reεults in a population of precursor pluripotent cellε with markedly reduced proliferative capacity. Finally, this results in CD4⁺ T-cellε that are relatively unreεponεive to stimuli to proliferate, as iε typical of the replicative εeneεcence of the cellε obεerved in vi tro .

We alεo propose that the replicative capacity of total peripheral lymphocytes or CD4⁺ cells in particular, can be effectively determined by aεεaying telomere repeat length utilizing the method deεcribed above, e.g. , with the oligonucleotide probe 5' TTAGGGTTAGGGTTAGGGTTAGGG (or one of similar or complementary sequence) hybridized to CD4⁺ lymphocyte DNA iεolated from the patient along with molecular size markers . These asεayε allow the phyεician to chart the courεe of the diεeaεe during the long intervening aεymptomatic period, and to score the effectiveness of prophylactic therapeutics .

In order to determine whether TRF length is a uεeful marker in diagnoεiε of HIV infection, CD4⁺ cell counting was performed on asymptomatic HIV-infected individualε, and compared to TRF length, meaεured as discuεεed above. As shown above, peripheral lymphocytes start with around 10 kb TRF length at birth, and reach a TRF length of 5.0 at approximately age 120. The results were as follows:

A 30 year old HIV+ with a CD4 count of 476 had a TRF of

7.6.

A 46 year old HIV- control, had a TRF of 7.0.

A 34 year old HIV+ with a CD4 count of 336, had a TRF of 7.7.

A 46 year old HIV- control, had a TRF of 7.1.

A 32 year old HIV+ with a CD4 count of 448, had a TRF of 6.9.

A 33 year old HIV+ with a CD4 count of 358, had a TRF of 5.0 (i.e. , at a length observed for seneεcent cells)

The results indicate that the 33 year old HIV+ patient has a εeneεcent telomere length in hiε CD4⁺ cellε, which meanε that they are at the end of their replicative capacity. In contrast, the CD4⁺ count provided no indication of the statuε of thiε patient. Indeed, one patient actually had a lower CD4⁺ count.

Two weeks after the asεay waε performed, thiε patient experienced a precipitous drop in CD4⁺ count, going from 358 to 159, and waε therefore diagnoεed AIDS, and rapidly acquired leukoplakia on the tongue. The other patients remain asymptomatic. Thuε, thiε diagnoεtic procedure is able to diεtinguiεh patientε near the end of the course of HIV infection, whereaε the previouεly used marker (CD4⁺ count) could not.

Referring to Fig. 42, it can be seen that terminal AIDS patients have a statiεtically significant decrease in TRF length in CD4, CD8 and total peripheral blood l mphocyteε compared to age-matched controls, almost to the extent that the TRF is close to that of centenarians.

The accelerated replicative seneεcence of CD4⁺ lymphocytes during the course of HIV infection provides an appropriate indication for therapies designed to forestall telomere shortening, e.g. , utilizing the CTO oligonucleotide described above. In addition, as described above, CD4⁺ cellε of an individual at an early εtage of infection can be banked for later administration to the individual . The efficacy of drugs, such aε AZT, may also be determined to study whether the drug slowε the rate of proliferation of CD4⁺ cells, and is thus useful at all εtageε of the diεeaεe. If not, it can be administered only when necesεary during the courεe of the diεeaεe.

Example 9 : Telomere Shortening in Human Mammary Epithelial (HME) Cells

Referring to Fig. 12, when digested with a restriction enzyme having a 4-baεe recognition site (like Hinfl) , most genomic DNA iε digested into small fragments. However, because the repetitive telomeric sequences lack reεtriction sites, telomeres retain relatively large terminal reεtriction fragmentε (TRFε) compoεed of 2-5 Kb of εubtelomeric DNA and age- dependent amounts of telomeric repeats . As previously described for human fibroblasts, lymphocytes and endothelial cells, telomere length εhortenε in normal human mammary epithelial cellε during in vi tro cellular εeneεcence (compare TRF length in laneε 1 (PDL 21) and 2 (PDL 40) ) . In human mammary epithelial cells expresεing E6 of human papilloma viruε 16, the TRF length continues to shorten during the extended lifespan period until crisis and subsequently immortalization occurs (lane 3 (PDL 68) ) . The TRFs generally stabilize in immortalized cells (lane 4 (PDL 81) and lane 5 (PDL 107) ) conεiεtent with the re- expression of telomerase activity.

Example 10 : Slowing Telomere Losε in Mammary Epithelial Cellε Reεultε in Increased Replicative Lifeεpan Normal human mammary epithelial cellε can be eεtabliεhed from organoidε (obtained from reduction mammoplasty) and can be cultured in defined condition in a standard medium (MCDB170) devoid of serum. Epithelial cellε with typical cobblestone morphology spread around organoids plated in thiε medium. After the first subcultivation theεe cultureε enter a period of growth arrest for 2-3 weeks until a population of small, highly birefringent and rapidly dividing cells expand among larger cells. The medium (MCDB 104) apparently selects for a lesε differentiated cell type with increased growth potential. These cells can be subcultured for 40-45 additional doubling before undergoing cellular εeneεcence.

Aε in Example 1, the change in proliferative lifespan and rate of telomere shortening in cultured mammary epithelial cells treated with the indicated amounts of CTO (occasionally referred to as C-Rich Terminal Repeat (CTR) ) versuε control random oligonucleotideε . Normal human mammary epithelial cellε from a donor (31) were infected with the E6 gene of human papilloma virus 16. This gene product bindε p53 protein and permitε HME31 cellε to have extended life εpan by proliferating from PDL 42 to PDL 62 when crisis occurs . During thiε extended lifeεpan period the TRFε εhorten from an average of approximately 5 kb to 2.5 kb (compare in Fig. 12 HME31 PD 40 to HME31E6 PD 68) . As is demonstrated in Fig. 13, experiments initiated using HME31E6 cells at PDL 36 were cultured in the preεence of 3, 10, 30 and 100 μM CTO. Aε controlε the cellε were cultured without oligonucleotides (nil) or with 30 μM random oligonucleotide. Fig. 13 demonstrates that compared to the nil control and the 30 μM random oligonucleotide, there was a dose related retardation of TRF shortening between PDL 36 and 50. This is most easily seen by examining the subpopulation of telomere TRFs that migrate more slowly than the rest, giving a discrete trailing band. Cells were maintained in logarithmic growth with medium changed and fresh oligonucleotide added three ti eε per week. Human mammary epithelial cells expressing

HPV16 E6 bypass Ml and have extended replicative lifespan. HME31 cells normally senesce at PDL 42-45. When expressing E6 they will bypasε Ml and divide until they reach criεis (M2) at PDL 53-62. The TRFs in HME31 (E6) cells at PDL 40 are approximately 5-6 Kb while at PDL 62 they are 3-4 Kb (see Figure 12) . As is demonεtrated in Figure 17, experimentε initiated uεing HME31E6 cells at PDL 36 were cultured in the presence of 30 μM and lOOμM CTR in defined medium without serum. As controlε, the cellε were cultured without oligonucleotide (control) , or with a 30 μM random oligonucleotide with the baεe content matched to the CTR oligonucleotide. Figure 17 demonstrates that compared to the control and the 30μM random oligonucleotide, there was a dose-related extension of the replicative lifeεpan in cellε treated with CTR oligonucleotides . The control cells divided approximately 20 times during the experiment, whereas the CTR-treated cells divided at least 40-50 times. These reεultε correlate well with the retardation of telomere εhortening obεerved in Fig. 13. Example 11: Extenεion of Life Span of IMR90 Fibroblasts

Referring to Fig. 14, IMR-90 lung fibroblasts TRFat PDL 30 were treated with 10 μM, 30 μM or 100 μM phosphodieεter CTO or with only media addition (control) . The cellε were cultured in medium containing regular defined εupplemented calf serum. The cellε were passaged in 24 well diεheε and εubcultivated by trypεinization upon reaching confluency at 25,000 cellε per well. The cellε were fed medium containing oligonucleotides at various concentrations daiiy. As a control, cells were fed identical medium without oligonucleotides. As is illustrated in Fig. 14, there was approximately a 12-15% extension of total life span with CTO. In these experiments the control cells divided approximately 15-18 timeε during the experiment, whereaε the treated cellε divided 23-26 timeε. IMR-90 telomeres shorten approximately 50 b.p. per division and the TRF length of the control IMR-90 fibroblasts at senescence was approximately 9 kb. Since the 100 μM CTO-treated IMR-90 cells seneεced at PDL 55, the predicted difference in the rate of TRF loεε between the control and the 100 μM CTO (9 kb vε 9.4 kb) is too small to be resolved uεing current techniqueε .

Example 12 : GTO Experimentε

Aε in Example 2, an immortalized human fibroblast cell line, IDH4, which has very short TRFε, waε incubated with GTO oligonucleotide. Referring to Figε . 15 and 16, cellε were incubated in regular culture medium containing serum in the presence of 10 μM, 30 μM and 100 μM GTO. The cells were fed fresh phoεphodieεter GTO oligonucleotide every other day and εubcultured when confluent for a total of 90 days. The cells were still growing in GTO after 90 days at all concentrations used even though they grew more slowly at the higher GTO concentrationε and went through fewer population doubling (control, 45 PDL; 10 μM GTO 40 PDL; 30 μM 35 PDL; 100 μM 25 PDL) . When TRF analysis was performed after 90 dayε the IDH4 cellε regained TRF length in a dose dependent manner with 30 μM and 100 μM being approximately the same (Fig. 15) . This suggests that the presence of exceεε single-stranded TTAGGG DNA in the cell was probably influencing the feedback regulation of telomerase and actually increasing telomerase activity and extending telomere length. The control and 30 μM GTO were passaged without oligonucleotide addition for an additional 90 days (approximately 35-40 PDL) . As is illustrated in Fig. 16, the TRFs slowly shorten.

Theεe data and those in Example 2, indicate that cell lineε differ in their response to GTO oligonucleotide. Thus, prior to use of εuch an oligonucleotide in therapeutic compositions it is important to ensure that the target cells respond aε desired. Should the effect seen above occur, then the oligonucleotide should be chosen to change the reεponse to that shown in Example 2. This can be done by choosing an oligonucleotide which binds to telomerase at a different site from that bound by GTO. Applicant believeε that the effect obεerved above is caused by binding of GTO to required proteins, allowing telomerase to be active to expand the telomeres. Thus, by choosing an oligonucleotide which does not bind such proteins the desired effect of reducing telomerase activity can be achieved. Example 13 : Small Molecule Inhibition of Telomerase

The following is an example of a method for screening for activity of small molecules aε inhibitorε of telomerase. Similar examples will be evident to those in the art. Compoundε that can be screened include those which are not thought to be cytotoxic because they do not cauεe immediate cell death. Rather, εuch compoundε act only after εeveral generations of inhibition of telomerase activity. Thuε, previouε drugs tested by standard means should now be reteεted to determine their utility aε claimed herein. Drugε which inhibit telomeraεe activity, or in εome caεes activate it in vivo (e.g.. at the level of transcription) are useful in treatment of disease are discussed herein.

We analyzed the effects of various nucleoεide analogε, which are chain-terminating inhibitors of retroviral reverse transcriptases, on Tetrahymena thermophila telomeraεe activity in vi tro, and on telomere length and maintenance, cell diviεion and conjugation of Tetrahymena cells in vivo . In vi tro aεεayε of telomeraεe activity εhowed that arabinofuranyl-guanoεine triphoεphate (Ara-GTP) and ddGTP were both very efficient inhibitorε of incorporation of labeled nucleotideε into telomeric DNA repeats, even at low inhibitor concentrations, while azidothymidine triphoεphate (AZT-TP) , dideoxyinoεine triphosphate (ddlTP) or ddTTP were lesε efficient inhibitors of incorporation. All of these nucleoside triphosphate analogs, however, produced analog-specific alterations of the normal banding patterns seen upon gel electrophoresis of the syntheεiε products of telomerase, suggesting that the competitive and/or chain terminating action differed at different positions along the RNA template. The effectε of theεe analogε in nucleoεide form on

Tetrahymena cell growth, conjugation, and telomere length were tested. Although cell division rates and viability were unaffected after several weeks in culture with Ara-G, telomeres were consistently and rapidly shortened in cultureε containing AZT or Ara-G, and growth rateε and viability of a fraction of cellε were decreaεed in AZT. In εhort-term experimentε with cultureε containing ddG, ddI,or 3' deoxy-2' , 3 ' -didehydrothymidine (d4T) , d4T also showed shortened telomereε. ddG or ddl had no effect on telomere length. AZT, Ara-G, Acycloguanoεine (Acyclo-G) , ddG and ddl were added to conjugating cells, but none εhowed any irreverεible disruption of conjugation or macronuclear development, as εhown by quantitation of the efficiency of formation of progeny cellε. PCR analysis of DNA from cells mated in AZT did show a decrease in the formation of 11Kb rDNA, a marker for telomere addition during Macronuclear developement.

The following materialε and methodε were uεed to obtain these resultε:

Tetrahymena thermophila strains SB210(VI) and PB9R(II) , where numbers in parentheseε indicate mating type, were maintained as stocks at room temperature in 1% PPYS (1% proteose peptone (Difco), 0.1% yeast extract (Difco) and 0.0015% Sequestrine (Ciba-Geigy) ) . Stocks were passaged every three to four weeks.

For analysis of macronuclear DNA from cultures containing the nucleoside analog AZT (Sigma) , or controls lacking analog, at various timepoints during vegetative divisionε, cellε from stationary stock cultures were inoculated into 25ml thymine-deficient Iso-sensiteεt broth

('Iεobroth', Oxoid USA) in 250 ml flaεkε. Cultures were incubated at 30°C with shaking (100 rpm) for 48 hours. Cells were counted and plated at 1000 cells/1.5ml in 24-well plates (Falcon) and grown at 30°C, without shaking, for 48 hours. 5 μl of theεe log phase cells were used to inoculate 1 ml cultures (Iεobroth) containing varied concentrationε of nucleoside analog. Thereafter, every 2-4 days cellε were transferred, either 5 μl per well, or 1-3 μl using a multi-pronged replicator into freεh 1 ml broth containing AZT. Remaining cells were pelleted and stored at - 80°C until processed for DNA analysis. For analysiε of macronuclear DNA from vegetative cultureε containing the nucleoεide analogε Ara-G (Calbiochem) , ddG (Calbiochem) , or ddl (Calbiochem) , or controlε lacking analog, εtock cultureε were grown overnight in 2% PPYS aε described. Cells were counted and plated at 100 cells/2 ml in 2% PPYS containing varied amounts of analog, 1% DMSO (Fisher) (as a control for ddG and Ara-G) , or 2% PPYS alone. Cells were replica plated into freεh medium every 2-6 dayε, and remaining cells were pelleted and stored at -80°C until procesεed for DNA analyεiε.

For analysis of macronuclear DNA from vegetative cultures containing d4T (Sigma) or control lacking the analog, stock cultureε (SB210 VI) were grown overnight in Isobroth as deεcribed. Cellε were then counted and duplicate cultureε inoculated at 500 cellε/5 ml Isobroth in 50 ml conical tubes, and grown at 30°C, shaking 80 rpm. 500-2000 cellε were tranεfered to fresh broth every 2-4 days, and the remainder pelleted and stored at -80°C until processed for DNA analysis .

For analysiε of rDNA from cells conjugated in the preεence of nucleoεide analogε, 50 ml overnight cultures (2% PPYS) were starved by pelleting cells and resuspending in an equal volume of Dryl's solution before returning to 30°C shaking (100 rpm) incubator for 18 hours. (IX Dryl's εolution = 0.5g Na citrate, 0.16g NaH₂P0₄ ^*H₂0, 0.14g Na₂HP0₄ per liter, plus 15 ml of 9.98g CaCl₂2H₂O/500ml) . Cells were then counted and equal numbers mixed before pelleting (6 minutes in an IEC tabletop centrifuge, 3/4 speed) , and resuεpended in Dryl's to 1.5-2xl0⁶/ml. Cells were plated at an average density of 1.5 cells/well into 6-well plates (Falcon) and allowed to conjugate 6 hours, 30°C without εhaking. Mock-conjugated SB210 cellε were treated identically but not mixed with PB9R cellε. At 6 hourε the cultures were checked for pairing (>90%, except SB210 controls) and either 1ml Dryl's εolution or 2% PPYS containing the nucleoεide analog (Acyclo-G purchaεed from Sigma) or no added drug aε control were added εlowly with gentle swirling. Cultures were returned to 30°C for an additional 18 hours before being harveεted for DNA analyεiε.

For analyεiε of vegetative growth and macronuclear

DNA from εingle-cell cultureε containing the nucleoεide analogε AZT or Ara-G, SB210 (VI) cells were grown from stationary stock cultureε overnight at 30°C with εhaking (100 rpm) in 50 ml 2% PPYS or Isobroth. Cells were counted and added to the appropriate medium plus analog (Ara-G to 1 mM or DMSO to 1% as control in 2% PPYS; AZT to 10 μM or 1 mM, or no addition as control in Iεobroth) and plated in 96-well plateε (Falcon) , 100 μl per well at a denεity of 1 cell per well. 5 plateε were prepared for each analog or control. Wellε were εcored for cell growth and plates were replica plated every 1-2 days (Ara-G and DMSO plateε) or every 2-4 days (AZT and Isobroth control plateε) to maintain approximate inoculation denεitieε of 1-10 cellε per well for each passage. Occasionally individual wells were passaged by hand (1 μl inoculated per well using a pipettor) into several blank wells, to expand the number of live wells per plate as single-cell cultures were lost over time due to low probability of being transferred at each passage. After passaging, cells were pooled, pelleted and stored at -80°C until procesεed for DNA analyεis.

Total cellular DNA was prepared essentially as deεcribed by Larεon 50 Cell. 477, 1987, except that the Hoechst 33258-CsCl gradient purification εtep waε omitted. Restriction digests, agarose gel electrophoresis, transfer of DNA to Nytran filters (Schleicher and Schuell) , and hybridization with ³²P-nick-translated or random-primed probes were carried out using standard procedures (Maniatis et.al. 1989). Telomere length waε analyzed aε deεcribed previouεly for Tetrahymena [Larεon 50 Cell, 477, 1987] . For analyεiε of cycloheximide (CHX) sensitivity of cells conjugated in the preεence of analog, 50 ml cultures of each cell type were grown overnight in 2% PPYS, starved in Dryl's for 18 hourε, mated (5xl0⁵ cellε/ml) for 6 hourε, then analog was added. Cells were allowed to complete mating in the presence of the analog. Twenty-four hourε after mixing, cellε were diluted in Dryl'ε εolution, counted and plated at 1 cell per well of 96-well plates in 1% PPYS without analog. Cells were grown for 4 days in a humid chamber at 30°C, without εhaking. Cells were then replica plated into 1% PPYS plus 15 μg/ml cycloheximide, allowed to grow for four dayε before εcoring, and percent of CHX- reεiεtant wells was calculated. Becauεe generation of progeny expresεing the cycloheximide marker requires successful production of a new macronucleus, cells whose macronuclear development was diεrupted by the analog are killed in CHX. For PCR analyεiε of the llkb form of the rDNA from cultures conjugated in the presence of analog, 1.25 μM each of the telomeric primer (C₄A₂)₄ and a 25-mer rDNA primer (5' GTGGCTTCACACAAAATCTAAGCGC 3') located 1371 nucleotideε from the 5' end of the rDNA were uεed in a "hot εtart" reaction containing 1 mM MgCl_2ι 0.2 mM each dNTP, IX PCR reaction buffer (Perkin Elmer Cetus) , and 0.5 μl Amplitaq polymerase (Perkin Elmer Cetus) . Sample DNA and polymerase were kept separate by the use of Ampliwax PCR Gem 100 wax beads (Perkin Elmer Cetuε) , following manufacturer's instructionε. The εamples were heated to 95° C for 1 minute, and then cycled 40 rounds in a Perkin-Elmer thermocycler aε followε: 1 minute at 94° C, 30 εecondε at 58° C, 3 minutes at 68° C. Identical reactions were done using 3' micronuclear rDNA primerε, 9610 nucleotideε from the 5'end, and

(5' CAATAATGTATTAAAAATATGCTACTTATGCATTATC 3') , 10300 nucleotideε from the 5' end.

Synthetic oligomers were prepared as described Greider 43 Cell, 405, 1985. Extractε were prepared aε described by Blackburn et.al. , 31 Genome 553, 1989.

A εtandard aεsay contained 50% by volume of heparin-agarose purified telomerase, 25 μM TTP, 1.25 μM ³²P- labeled dGTP (400 Ci/ mMol, Amersham) , lμM oligo (either

(T₂G₄)₄ or (T₂G₄)₂ mixed with water and heated at 90°C for two minuteε and cooled at 30°C for 10 minuteε) , and 0.lμl RNasin

(40 U/ml, Promega) in a no-salt buffer. AZT-triphosphate waε obtained from Burroughε Wellcome, N.C. Ara-G- triphoεphate waε purchaεed from Calbiochem and ddNTPs from

Sigma. Reaction mixes were kept on ice until ready for use, and then mixed into tubes containing analog for incubation at 30°C. Reaction times were thirty minuteε. Reaction rateε under these conditions were determined previouεly to be linear over time for thirty minutes. Identical reactions were run without primers as controls. The reactions were then processed esεentially as described by Greider and Blackburn 337 Nature, 331, 1989. For quantitative asεayε, aliquots of the reaction mixture were spotted in triplicate onto DE81 paper and washed as described Greider 43 Cell, 405, 1985. Incorporation of ³²P label from either ³²p-TTP or ³²P-dGTP was measured to monitor the reaction rate. For visualization of the elongation reaction products, samples were heated to 95°C for 2 minutes and cooled on ice before loading onto a 12% polyacrylamide/8 M urea gel.

The model for the mechanism of the telomeraεe ribonucleoprotein enzyme from Tetrahymena is shown in Figure 18A. The enzyme syntheεizeε TTGGGG repeatε onto the 3' end of a εuitable DNA primer by copying a template εequence in the RNA moiety of the enzyme. For ease of reference in discussing the resultε, the reεidueε in the template region are numbered 1 to 9 (5' to 3' along the RNA) . The εtandard telomeraεe assay used in this example conεiεtε of incorporation of dGTP and TTP substrates, one triphoεphate ³²P-labeled, into εynthesized DNA in the reaction shown in Figure 18A. For the experiments discuεεed in thiε example we uεed aε the DNA primer either 1 μM (T₂G₄)₄ or (T₂G₄)₂, under conditionε in which the overall rate of incorporation of label was determined previously to be linear over time. Incorporation of ³ P label from either ³²p-TTP or ³²P-dGTP was measured to monitor the reaction rate, and the distributions of elongation productε were analyzed by denaturing polyacrylamide gel electrophoreεiε.

The effect of adding increaεing amountε of AZT- triphoεphate (AZT-TP) to the standard assay for telomeraεe activity iε shown in Figure 19A. A serieε of control reactionε using unlabeled TTP added at the same concentrationε aε the AZT-TP waε run in parallel (Figure 19A) . The unlabeled TTP inhibitε incorporation of the ³²P- labeled TTP by simple competition. Quantitation of label incorporated into product in this experiment enabled us to determine the K-. for TTP to be ~5 μM. Compared with addition of unlabeled TTP competitor, AZT-TP had only a modest quantitative effect on the incorporation of ³²P-labeled TTP (Figure 19A) . Since AZT incorporation leads to chain termination, this result indicates that AZT-triphoεphate competeε leεs efficiently for telomerase than TTP. Similar results were obtained when incorporation of ³²P-dGTP was monitored (Figure 19B) , with 50% inhibition occurring at ~80 μM AZT-TP.

In similar experimentε in which increaεing concentrations of arabinofuranyl-guanosine triphosphate (Ara-GTP) were added to the reaction, significant reduction of overall incorporation occurred even at low concentrations of the analog (Figure 19C) . From parallel experiments in which unlabeled dGTPwas added as competitor (Figure 19C) , the I - for dGTP under these reaction conditions was found to be 1-2 μM. 50% inhibition occurred with 0.7 μM Ara-GTP; thuε Ara-GTP potentially competeε as well as unlabeled dGTP for ³²P-dGTP. However, aε incorporation of Ara-G cauεeε chain termination, each Ara-G incorporated iε expected to have a greater impact on total incorporation than competition with unlabeled dGTP.

We alεo teεted the effectε of dideoxynucleoside triphosphateε (ddNTPs) on the telomerase reaction. As shown previouεly for telomeraεe [Greider 43 Cell, 405, 1985] , and aε is the case for many other reverse trancriptases, ddNTPε are recognized by the enzyme and incorporated, causing chain termination with a subsequent shift in banding patterns and reduction of the average product length. Consiεtent with previouε qualitative analyεeε of Tetrahymena and human telomeraεeε [Greider 43 Cell, 405, 1985; Morin 59 Cell. 521, 1989] , ddGTP and ddTTP each inhibited the incorporation of labeled ³ P-NTP into elongation productε (Figure 19D and E) . ddGTP was a much more efficient inhibitor than ddTTP: under these reaction conditionε 50% inhibition occurred at < 0.1 and 5 μM ddGTP and ddTTP reεpectively. Aε obεerved previously for Tetrahymena telomerase [Greider 43 Cell, 405, 1985] , no significant effects were seen with either ddCTP or ddATP. In addition, ddlTP inhibited telomerase (Figure 19E) , although lesε efficiently than ddGTP, with 50% inhibition occurring at 3 μM ddlTP.

The εize distribution of labeled products was then analyzed by denaturing polyacrylamide gel electrophoresiε . Consiεtent with the expectation for a chain-terminator, the proportion of longer telomerase products was decreased in the presence of AZT compared with cold TTP competitor controls (Figure 20A; compare lanes 1 and 2 with lanes 3 to 5) , and in the presence of Ara-G (Figure 20A; lanes 7 and 8) . Average product length also decreased in the preεence of Ara-GTP, ddGTP and ddlTP (Figure 20A and B) . In addition, each nucleoside triphosphate analog produced distinctive and characteristic patterns of chain termination, aε εhown by analysis of the shifts in the banding patterns of the elongation products. With AZT- triphosphate, we saw increased relative intensities of the bands corresponding to the incorporation of T reεidueε (copying the A reεidues at poεitionε 2 and 3 on the template RNA (εee Figure 18A) ) . This change in banding pattern is consistent with simple chain termination, which iε predicted to increaεe the intenεity of bandε corresponding to the position of both incorporated T residues. Similar effects were seen with ddTTP. We interpret thiε to mean that AZT- triphosphate was recognized by the enzyme and incorporated into the correct poεitionε in the growing telomeric sequence, causing chain termination. However it cannot be excluded that the increase in relative intensity of the band corresponding to position 3 on the template, which precedes addition of the second T, iε also attributable to pausing cauεed by competition with TTP and a εlower reaction rate with AZT-triphoεphate at poεition 2. Should AZT- triphoεphate, or related nucleotide analogε, be incorporated into telomeric DNA where they would not be incorporated by DNA polymeraεe into other DNA, then εuch nucleotide analogε may be uεed to kill telomerase positive cells by cauεing them to generate telomeric DNA toxic to the cell, or at least altered in such a way that telomeraεe-mediated cell immortalization was inhibited. The results with Ara-GTP were also consistent with incorporation of Ara-G and consequent chain termination (Figure 20A, lanes 7 and 8) . Although there are four poεitionε at which a G reεidue can be incorporated and therefore at which chain termination could occur, the εtrongeεt increase was in the band corresponding to the G residue εpecified by poεition 4, in the middle of the telomeraεe RNA template (εee Figure 18A) . With ddGTP, chain termination appeared to occur moεt efficiently at positions 6 and 5 (Figure 2OB, compare lane 1 with lanes 4 to 6) , and with ddlTP, at position 5 (lanes 7 to 9) .

Figure 18B summarizeε εchematically the effectε of the variouε triphoεphate analogε on polymerization at each of the εix poεitionε along the template. There waε no correlation between the efficiency of a nucleoside analog as an inhibitor and the poεition of itε maximal chain termination on the template. For example, the potent inhibitorε ddG-and Ara-G-triphosphates cauεe maximal chain termination at different poεitionε on the telomeraεe RNA template (5 and 6 for ddG, and 4 for Ara-G) . In addition to nucleoεide triphoεphate analogs expected to act aε chain terminators, we also tested rifampin, an inhibitor of bacterial RNA polymerase, and streptomycin sulfate. Streptomycin sulfate iε known to inhibit the activity of group I εelf-εplicing intronε at high concentrationε [von Ahεen 19 Nucl. Acids Res .. 2261, 1991] , and haε a guanidino group that might be recognized by telomerase as part of the enzyme's specificity for G-rich DNA primers (Greider 51 Cell, 887, 1987) . Adding rifampin at concentrations up to 100 μg/ml did not affect the quantitative incorporation of label or change the banding pattern of the elongation products . Streptomycin sulfate at 40 mM dramatically reduced the amount (Figure 19F) and average length of elongation productε, with little decrease in activity being seen in a 40 mM sodium sulfate control. However, unlike the nucleoside triphosphate analogs, inhibition by streptomycin did not appear to affect incorporation at specific positions in the repeat. The inhibition by streptomycin may be uεeful experimentally aε a criterion for telomeraεe activity in vitro . However, the εignificance of the inhibition by streptomycin is unclear, aε it iε difficult to rule out that itε effect iε the result of nonspecific binding to either the RNA moiety of telomerase or the DNA primer.

Because the triphosphate forms of the analogs AZT, Ara-G, ddT, ddG and ddl each inhibited (with varying efficiencies) telomeraεe in vi tro, we teεted whether supplying each of these nucleoside analogs in the cell growth medium caused in vivo changes in telomere length or senescence. Additionally, Acyclo-G and d4T were tested on conjugating and vegetative cells, respectively.

Previous work with Tetrahymena showed that at leaεt one alteration of the telomeraεe RNA cauεeε telomere shortening and cellular seneεcence [Yu 344 Nature 126, 1990] . To teεt whether εuch a phenotype could be produced by inhibitorε of telomeraεe in Tetrahymena, duplicate log- phase cultures were grown for prolonged periods in the presence of varying concentrationε of analogs. The growth and cell morphology of these cultures were monitored, and DNA was isolated at different times for telomere length analysiε. AZT at 5 or 10 mM added to Iεobroth medium εtrongly inhibited cell growth and killed cellε within a day, and thuε at these concentrations acted in a manner suggeεtive of immediate toxicity to cellε, rather than of εenescence. AZT added to Isobroth medium at lower concentrations (up to 1 mM) did not reεult in εeneεcence of cultureε maintained by εubculture of ~10³ cells per transfer, over a 50-day period of continuous growth and subculturing of these cell cultureε. From growth rate meaεurementε it waε calculated that the cellε went through 150 to 250 cell generations in the course of this 50 day period. In similar mass transfer experiments no effects on cell doubling rate, morphology or long term viability were obtained with cellε grown in 2% PPYS plus up to 2 mM Ara-G, the highest concentration tested that did not cause immediate toxicity.

Telomere lengths in cellε grown in the preεence of the different analogε were monitored by Southern blot analyεiε of DNA samples extracted at a series of time points during the subculturingε. The telomereε of cellε grown vegetatively in 1 and 5 mM AZT in 2% PPYS medium were reproducibly shortened by up to an average of 170 base pairε compared with the control cultures grown in 2% PPYS in the absence of the drug (Figure 21A and B) . This shortening of telomeres occured in a concentration-dependent manner (Figure 21B) , with at leaεt 50% of the maximal shortening effect occurring by 10 μM AZT, the lowest concentration teεted. For each AZT concentration tested, the full decrease was seen within 3 days of culturing in the presence of the drug (15 to 30 cell diviεionε) , but after this initial length adjustment, at each drug concentration telomeres thereafter showed no statistically significant shortening over time, and mean telomere length consistently remained static for at least 28 days of mass transfer εubculturing. Similar degreeε and timing of telomere εhortening were produced with 1 or 2 mM Ara-G added to 2% PPYS culture medium (Figure 21C) . d4T added to Iεobroth culture medium in concentrationε ranging from 10 μM to 1 mM produced shortened telomeres at 100 μM and 1 mM, again in a concentration dependent manner, after 5 days (16 generations) in culture. In contraεt, up to 1 mM ddG or ddl produced no changeε in telomere length compared with control cultureε, over a period of 5 days of subculturing (15-20 cell generationε) in 2% PPYS medium. Because we had found previously that telomeraεe is strongly inhibited in vi tro by at leaεt εome of the analogs tested, and telomere length is affected in vivo within an eεtimated 15 to 30 cell generations by these analogs, it was posεible that telomere addition waε in fact being disrupted in vivo, but that our failure to find any evidence of progressive telomere shortening or seneεcence was attributable to a subset of the cell population that escapes an inhibitory effect of the analog on telomerase. We have shown previously that impairing telomerase in vivo by mutating the telomerase RNA produced senescence in most cellε, but only -10 ² εingle cell εubcloneε were analyzed in these experiments, [Yu 344 Nature 126, 1990] . Under our mass transfer subculturing regime, in which about 10³ cells were transferred per passage, if a fraction as small as -1% of the cellε escaped senescence, and if their growth advantage was sufficiently high compared with cells losing telomeres, they could become the predominant population in any cell passage and we would not have detected any phenotype.

To test whether we had miεεed such a subpopulation of cells, we carried out the same experimentε on vegetatively dividing Tetrahymena cellε in the preεence and abεence of drug, but in these experiments the subculturing was carried out by plating cells at an average of 1 to 10 cells per well in microtiter plate wells in the preεence of 10 μM and 1 mM AZT, and 10 μM and 1 mM Ara-G. For each drug, cells were plated out in this manner for 30 consecutive days (90 to 150 cell generations) and 16 consecutive dayε (50 to 80 cell generations) respectively for the 10 μM and 1 mM drug concentrations. DNA was isolated at intervals from combined samplings of the wellε for analyεiε of telomere length.

Compared with control medium lacking the nucleoεide analog, no changeε in the plating efficiency were observed over the course of the experiment for cells grown in 10 μM AZT and 10 μM or 1 mM Ara-G. However, in the presence of 1 mM AZT, monitoring growth rates of cells maintained in this way by single cell transferε allowed uε to identify two general growth classes, which we designated as slow (0 to 1 cell doubling per day) and faεt (2 to 4 cell doubling per day) . The growth rate of fast cells was similar to that of the controls grown in Isobroth containing no AZT. Over time, the proportion of wells with slow cells decreased, as would be expected if they simply had a lower probability of being transferred, since they were preεent in lower cell densities than fast cells, which grew to higher cell denεitieε and for which the timing of the plating protocols had been worked out. However, monitoring the cells remaining in wells after tranεferε had been made from them showed that the slow cellε loεt viability over time. In addition, throughout the courεe of the tranεferε, εlow cells appeared from formerly fast cell wells. We pooled cellε from the εlow growing wellε (pooling of εeveral microtiter wellε waε neceεεary to obtain sufficient DNA for Southern analysiε) and compared their telomere length diεtribution with that of pooled faεt cellε . The mean length and εize distribution of telomeric DNA from pooled faεt cells were indistinguiεhable from those of control cells grown without AZT. In contrast, the pooled slow cell DNA showed a slight decrease in mean telomere length and heterogeneity (Figure 21D) . Control cells grown in Isobroth medium had telomeres that were an average of 165 bp shorter than cells grown in 1% PPYS medium. We believe that because the telomeric G₄T₂ repeat tractε in cells grown in Isobroth medium are already markedly shorter than thoεe of cells grown in the richer PPYS medium, the additional amount of telomere shortening caused by growth in 1 mM AZT iε sufficient to reduce continually and stochaεtically a fraction of the telomereε below a critical lower threεhold required for function, thus causing the decreased viability of a subpopulation of the cells.

We examined the effects of AZT, Ara-G, Acyclo-G, ddl and ddG on progeny formation by cellε that have undergone conjugation. Thiε proceεε involveε de novo formation of new macronuclear telomereε in the progeny cellε. Macronuclear development in ciliated protozoans such as Tetrahymena involveε developmentally programmed, site- εpecific fragmentation of germline chromoεomeε into linear εubchromosomes, whose ends are healed by de novo addition of telomeres. We showed previously that telomerase not only elongates pre-existing telomereε in vivo during vegetative cell divisions [Yu 344 Nature, 126, 1990] , but also functionε to directly add telomeric DNA onto non- telomeric εequenceε during this developmentally-controlled chromosome healing. Becauεe of the immediate requirement for telomere addition to fragmented DNA, it iε possible that the latter process might be more sensitive to telomerase inhibition than telomere maintenance during vegetative growth. To test whether nucleoεide analogε cauεe inhibition of macronuclear development due to a diεruption of telomere formation, we mated two strains of Tetrahymena which are senεitive to cycloheximide, but whoεe progeny after mating are resistant to cycloheximide. Synchronized mated cells were treated with AZT at concentrations ranging from 10 μM to 5 mM for a period beginning just prior to when macronuclear development beginε and continuing during macronuclear development (the period 6 - 24 hourε after mating waε initiated) . At thiε point cellε were diluted out in microtiter plate wells in freεh medium lacking the analog, at an average cell denεity of one cell per well, and allowed to grow for the minimum period before εelection for cellε that had εuccessfully produced progeny. In attempts to maximize the effect of AZT, cells were either refed at 6 hrs with 2% PPYS or Iεobroth, or εtarved until 24 hrs (the duration of the AZT treatment) . Such starvation arrestε macronuclear development at an intermediate stage. When refed, macronuclear development would then be forced to proceed in the preεence of the AZT. Control, unmated parental cellε were alεo plated and expoεed to drug. Similar experiments were performed with Ara-G, Acyclo-G, ddl and ddG. The resultε are εhown in Table 4.

The control plateε showed 99%-100% cell death in CHX, while the majority of cells that were mated with or without analog survived. None of the nucleoside analogs had any statistically significant effect on progeny formation. The design of thiε experiment would prevent takeover of the culture by a minority population that evaded the effectε of the drug, aε deεcribed above. Therefore little or no irreversible disruption of macronuclear development due to impaired telomerase activity and telomere formation occurred in the preεence of AZT, Ara-G, Acyclo-G, ddG, or ddl.

Although macronuclear development was not significantly disrupted, analysis of the formation of a marker for telomere addition during macronuclear development suggeεtε that AZT reduceε the efficiency of telomere addition.

DNA from cellε mated in the preεence or abεence of analog, and either refed at 6 hourε or εtarved fully for the duration of conjugation were uεed in PCR with a telomeric primer and a 5' rDNA primer. Thiε selected for a fragment of the llkb rDNA to which telomeres had been added. The 11 kb rDNA iε either a by-product of the 21 kb rDNA formed during macronuclear development or an intermediate of thiε proceεε. It iε preεent only tranεiently during new macronuclear development and aε εuch iε a good marker for telomere addition in vivo . Knock-down of relative amounts of the 1400 nucleotide PCR-generated fragment from 11 kb- rDNA was seen in DNA from cells conjugated in the presence of AZT, but not in thoεe containing Ara-G, Acyclo- G, H₂0 or DMSO controlε or in mock-conjugated SB210 cellε. To show that the DNA uεed in the PCR reactionε was present and competent for PCR, identical reactions were run using primers from the 3' -micronuclear copy of the rDNA. In all sampleε the expected 810 nucleotide fragment waε produced in εubεtantial quantitieε (fig. 22) , indicating that the decreaεe in the 1400 nucleotide telomere-containing PCR product in εampleε from cells mated in AZT is due to the presence of analog rather than contaminants in the DNA or reagents. Southern blotting with a 5'-rDNA probe confirmed that the telomere-containing PCR product was from the expected rDNA εequence, (figure 22B) and no cross- hybridization occurred to the 3' PCR product. An overall decrease in telomere-containing PCR products was seen in all samples that were re-fed at 6 hourε poεt-mixing, but the decreaεe was more pronounced in samples that had been mated in the presence of AZT.

TABLE 6

Effects of nucleoside analogs on progeny formation.

CELL TREATMENT # CHX-R # TOTAL %3K-R

SB210 (NOT MATED) 1 215 0.5

PB9R (NOT MATED) 3 307 1

AZT (mM) 0 139 212 66

0.01 121 169 72

0.1 100 148 68

1.0 91 166 55

5.0 75 120 63

SB210 (NOT MATED) 0 57 0

AZT (mM) 0 165 214 77

0.01 67 92 73

0.1 128 190 67

1.0 60 125 48

5.0 89 168 53

1% DMSO 84 109 77

ARA-G (mM) 0.01 114 141 81

0.1 134 167 80

1.0 89 161 55

DMSO 51 75 68 ARA-G (mM)l.O 51 86 59

2.0 40 92 43

SB210 (NOT MATED) 0 9 0

PB9R (NOT MATED) 0 37 0 ddl (mM) 0 63 75 84

0.001 59 71 83

0.01 85 96 89 0 0..11 8 833 106 78

1.0 100 110 91

1% DMSO 21 44 48 ddG (mM) 0.001 86 102 84

0.1 73 86 85 1 1..00 5 511 66 77

ACYCLO-G (mM) 0 36 45 80

0.017 80 107 75

0 78 116 67 0 0..001177 1 10011 146 69

Example 14 : G-Reaction for Reducing the Size of the Terminal Reεtriction Fragment Human fibroblaεt DNA digeεted with reεtriction enzymeε, electrophoreεed, and hybridized by Southern blot makes posεible the resolution of terminal reεtriction fragmentε (TRFs) which in turn reflect the relative length of telomeric repeat sequences (See Fig. 26, Hinfl digested DNA, labeled "Hinfl"; DNA not digested, labeled "0") . This Southern analysiε iε complicated by the fact that human and many other εpecieε have long εtretches of subtelomeric repetitive εequences that add to the TRF size. As a means of eliminating the artifactual inclusion of this subtelomeric repeatε in a meaεurement of telomeric repeat length, a modified Maxam-Gilbert reaction iε employed to hydrolyze the DNA at G residueε. In the lane labeled "P only" (underloaded) the DNA iε treated with piperidine in mild conditionε which doeε not in itεelf decreaεe the size of the DNA. In the lane labeled "P+DMS" the sampleε are pretreated with DMS. Not the εubεtantial reduction in TRF εize compared to the Hinfl digeεt relecting the deletion of subtelomeric sequences in the C-rich strand containing G reεidueε. All lanes were probed with (TTAGGG)₃. This aεεay iε thuε useful for analysiε of telomere lengths in diagnostic procedures. Example 15: Fungal telomeres

The following example illustrates various specific telomeric sequenceε which can be uεed to identify specific fungi. Those in the art will recognize that such sequences can be probed with oligonucleotides to specifically diagnose the presence of a selected fungus. In addition, specific treatment of fungi can be effected by uεe of agentε which bind to such sequences and reduce the long term viability of the fungal cell.

Aε deεcribed herein telomeric DNA iε an attractive target for εpecific drug therapy. Telomeres are short single-εtranded protruεionε which are acceεεible to εpecific drugε. Binding by εuch drugε will interfere with normal telomere function and thus fungal cell viability. In similar experiments (routine to those in the art when conducted as described herein) inhibitors or facilitators of such telomere replication (or telomeraεe activity) can be diεcovered and used as anticancer, antiparasite and antifungal agentε.

The εignificantly increaεed length of fungal telomereε makeε them ideal targets for antisenεe therapy or diagnosis. In addition, this different telomere structure indicates a different mechaniεm of action of the telomeraεe, and thuε itε availability aε a target for antifungal agents which are inactive on human or other animal cells.

Telomeric DNA sequences have generally been found to be remarkably conserved in evolution, typically consisting of repeated, very short sequence units containing clusters of G residues. Recently however the telomeric DNA of the budding yeast Candida albicanε was εhown to consist of much longer repeat units. Here we report the identification of seven additional new telomeric εequences from budding yeasts. Although within the budding yeastε the telomeric εequences show more phylogenetic diversity in length (8 - 25 bp) , sequence and composition than has been seen previously throughout the whole phylogenetic range of other eukaryotes, we show that all the known budding yeast telomeric repeats contain a strikingly conεerved 6 bp motif of T and G reεidueε resembling more typical telomeric sequences. We propose that G clusterε in telomereε are conεerved becauεe of conεtraintε imposed by their mode of syntheεis, rather than by a fundamental requirement for a specific common structural property of telomeric DNA.

The DNA εequences of telomeres, the ends of eukaryotic chromosomes, have been found previously to be conserved even between very diverse eukaryoteε, typically conεiεting of tandem arrayε of 5-8 bp repeating unitε characterized by cluεterε of G reεidueε, producing a marked strand composition bias. However, the telomeric repeats of the opportunistic pathogen Candida albicans were shown to consist of homogeneous repeats of a 23 bp sequence that lacks any noticeable strand compoεition bias.

To determine the relationship of the apparently exceptional, complex telomeric repeat sequence of Candida albicanε to the more uεual, εimple telomeric sequenceε, genomic DNA from budding yeast species related to both C. albicanε and S. cereviεiae were analyzed by Southern blotting, using cloned C. albicanε telomeric repeats as the hybridization probe. Under low-stringency hybridization conditions we detected multiple cross-hybridizing bands in several specieε Fig. 28. In some cases, the crosε-hybridizing bandε clearly were broad, a characteriεtic feature of telomeric reεtriction fragments caused by different numberε of telomeric repeatε in individual telomereε among a population of cellε.

Telomere-enriched libraries were constructed from genomic DNA from seven budding yeast species and strains. Telomeric clones were identified by their ability to hybridize to known yeast telomeric repeats (either the 23 bp C. albicanε repeat or the TG^ repeat of S. cerevisiae) , or by screening for end-linked repetitive DNA sequences without the use of a specific probe. Sequencing putative telomere fragment inserts from seven species identified clones that contained tandem repeats with unit lengths of 8-25 bp. With a εingle exception, the repeatε showed no sequence variations within a εpecieε. In every case the repeat array was present at the very end of the insert, directly abutting vector sequenceε, as would be expected for cloned telomeres. The repeat-containing clone from each specieε hybridized back to the εame pattern of reεtriction fragments observed originally with the C. albicanε or the S. cereviεiae probe used for library εcreening. Moεt of the bandε were preferentially sensitive to Bal31 nuclease (Fig. 29) indicating that the bulk of the repeat sequences are present at the ends of chromosomes. The lengths of the tracts of repeatε cloned from the different yeaεt εpecies were typically between 250-600 bp, although thoεe from the two C. tropicaliε strainε were only 130- 175 bp. That this species has particularly short telomeres iε alεo εupported by their very rapid loss during Bal31 digestion and by the relatively weak hybridization, even with species-εpecific telomere probes.

Figure 30 showε an alignment of theεe newly diεcovered telomeric repeat unit εequences together with those of C. albicanε and S. cerevisiae . Two εtriking featureε are apparent: the much greater variety of the budding yeaεt telomereε, with reεpect to repeat unit lengths and sequence complexities, compared to other eukaryotes, and a conserved six-base cluster of T and G residues that most reεembles typical telomeric sequences.

The sequence relationεhipε among the telomeric repeats are generally consistent with the phylogenetic relationshipε of theεe budding yeaεtε. The telomeric repeats of the two C. tropicalis strainε differ by only a single base polymorphism. The 25 bp telomeric repeats of the cloεely related K. lactis and C. pεeudotropicaliε differ at only one poεition. The telomeric repeat sequences from C. albicans, C. maltoεa, C. pεeudotropicaliε, C. tropicaliε and K. lactiε are 23-25bp in length, with differences largely or entirely confined to the central part of the repeat. The 16 bp repeat unit from C. glabrata, the species in this study that may be moεt closely related to S. cereviεiae, is very G-rich, which probably contributes to its croεε-hybridization to the heterogeneouε and εmaller S. cereviεiae telomeric repeats. All the budding yeast εequenceε, including the irregular S. cereviεiae repeatε, have a perfect or 5/6 match to a 6 bp T/G εequence (boxed) . In the cloned telomere from C. tropicaliε εtrain B-4414, we found two telomeric repeat εequenceε that differed at the second base position of the repeat, as εhown in Fig. 30 repeat units in the B-4414 telomere were homogeneous (and will be termed the "AC repeat"), but the remaining repeat (henceforth termed the "AA repeat") waε identical to the homogeneouε telomeric repeatε cloned from εtrain C. tropicaliε B- 4443.

To determine the distribution of these variant repeatε among the telomereε and εtrainε of C. tropicalis, genomic DNA from εeveral C. tropicalis εtrains including B-4414 and B-4443, and a control C. albicanε strain were probed with oligonucleotide probeε εpecific for either the AA or the AC repeat (Fig. 31 left panel) . Only εtrainε B-4414 and 1739-82, and to εome extent the C. albicanε telomeres, hybridized with the AC repeat-specific oligonucleotide probe (Fig. 31 left panel) . However, genomic DNA from all of the C. tropicaliε strainε teεted, including B-4443, but not from C. albicanε, hybridized well with the oligonucleotide specific for "AA" repeats (Fig. 31 right panel) . These resultε clearly indicate that both B-4414 and 1739-82 contain at leaεt two formε of telomeric repeatε, which are moεt likely variably interεpersed in different telomeres, as signal rations with the two probes differed between individual telomeric fragments (Figure 31A and B, laneε 1 and 2) . Example 16 : Effects of Telomerase Inhibitors on Human Tumor Cell Growth

Agentε that were εhown to inhibit telomerase from Tetrahymena e.g. , AZT, ddG, and ara-G were tested to determine their effect on human telomerase activity, telomere repeat length, and cell growth immortality. Of the compounds tested ddG and ara-G were effective inhibitorε of human telomerase obtained from the tumor cell line 296. The data for ddG is εhown in Fig. 27. The effect of the agentε on telomeraεe activity in intact cellε waε then εtudied utilizing the lymphoma cell line JY 616 which were maintained in RPMI 1640 with 0.25M Hepes, 10% FCS, and penicillin/streptomycin

(Gibco) . The cellε were cultured in 6-well plates (Falcon) with 5.0 ML of medium per well in duplicate. Cells were passaged every 7-10 days which corresponded to 5-7 mean population doublings (MPD) , and seeded at 3 x IO⁴ cells per well into fresh medium containing analog or control. Cell viability was monitored prior to harvest utilizing trypan blue stain (Gibco) during counting with a hemocytometer. The average ratio of stained: unstained cellε (dead:alive) waε > 90%. The intactness of the DNA was meaεured on a parallel gel by obεerving itε mobility in a gel prior to digestion by a restriction enzyme.

As seen in Fig. 23, all JY cells grew in an immortal fashion in the preεence of a low concentration of the potential telomeraεe inhibitorε. At high concentrationε (Fig. 24) , the cellε ceased proliferating in the presence of 50μM AZT and displayed a slowed growth in the preεence of 20μM ara-G. In εupport of the belief that thiε inhibition of cell growth in the preεence of 50μM AZT, iε due to telomerase inhibition, is the observation that the cells grew at a normal rate until week 3 and then ceased dividing. This iε the effect one would expect if the inhibition of cell growth was via telomerase inhibition (i.e. , the cells require multiple roundε of cell diviεion to loεe their telomeric repeats) . Also in support of the belief that AZT inhibited the growth of the cellε via the inhibition of telomeraεe iε the finding εhown in Fig. 25 where compared to week 1, and week 3 where the cells stopped dividing, the AZT treated cellε had a marked decrease in mean telomere length compared to the control medium "R" at the εame time. In addition, lOμM ddG waε shown to cause a decrease in telomere length compared to the control (in this caεe a DMSO control) . In Fig. 32 it can be εeen that JY cellε εtudied in a manner εimilar to that deεcribed above, and treated with ddG, εhowed a markedly εhorter telomere repeat length after 9 and 10 weekε compared to the DMSO control. It εhould be noted that while JY cells are immortal, when cultured under the conditions described, they lose some telomeric repeats over 10 weeks . The addition of ddG markedly accelerated this loss.

Example 17 : An Alternative Method of Measuring Telomere Repeat Length

An alternative method to measure telomere length exploits the fact that the telomere sequence lacks guanine reεidues in the C-rich strand. Unmelted genomic DNA can be mixed with a biotinylated oligonucleotide containing the sequence Biotinyl-X-CCCTAACCCTAA which will anneal to the single stranded G-rich overhang, followed by extension with the Klenow fragment of DNA polymerase in the presence of dTTP, dATP and radioactive dCTP. The DNA is then mixed with streptavidin-coated magnetic beads, and the DNA-biotin-streptavidin complexeε recovered with a magnet . Thiε procedure purifieε the telomeres and the radioactivity recovered at thiε step is proportional to the number of telomeres. The DNA is then melted, and DNA synthesis primed with fresh CCCTAACCCTAA oligonucleotide, dTTP, dATP and radioactive dCTP. The radioactivity incorporated during this εtep iε proportional to the number of telomeric repeatε

(telomere length) after correction for the number of telomereε present as determined during the first step. This value can then be converted into an actual telomere length by compariεon to a εtandard curve generated from telomereε of previouεly determined lengths .

Example 18: An Alternative Method to Isolate Telomeric Sequences

Large telomeric DNA is purified as follows . A biotinylated oligonucleotide with the sequence biotinyl-X-CCCTAACCCTAA is used to prime DNA synthesis in double-stranded DNA. The only sequences with which this oligonucleotide can anneal will be the single-stranded base overhangs aε telomere endε. The extended DNA, which now has a more stable structure than that provided by the initial 12 bp overlap, is then recovered using streptavidin. For large DNA, the DNA could be digested with a rare-cutting restriction endonuclease such as Notl, then εubjected to pulse-field electrophoresis, Streptavidin, covalently attached to a block of agarose near the origin, would bind to the biotinylated DNA and restrict the migration of the telomeres while permitting the bulk of genomic DNA to migrate into the gel. Telomeric DNA could then be recovered, cloned and characterized.

Alternately, smaller telomeric DNA fragments are recovered from sheared DNA using streptavidin coated magnetic beadε. The following method was used to obtain these results:

Pilot experiments had indicated that the shearing forces generated during the mixing and separation procedure yielded DNA fragmentε approximately 20 kbp long. In order to maximize the amount of εubtelomeric DNA obtained, DNA from a T-antigen immortalized cell line (IDH4, derived from IMR90 human lung fibroblasts) that had very few telomeric repeats (short TRFs) were used as the source of the DNA. 50μg of IDH4 DNA was mixed with 1.25 pmol of biotinylated CCCTAACCCTAA primer, 33μM each of dATP, dTTP and dCTP, and 2U of the Klenow fragment of DNA polymerase, in a final volume of lOOμl of Boehringer Mannheim restriction endonuclease Buffer A and extended for three hourε at 37°C. A εimilar amount of a biotinylated TTAGGGTTAGGG primer (which εhould not anneal to the G-rich telomeric overhang) waε added to a second reaction as a negative control. Five μl of M-280 Streptavidin-coated magnetic beads (Dynal, Inc.) were then added and gently mixed for 2 hours at room temperature, then biotinylated DNA-streptavidin-bead complexeε were recovered by holding a magnet against the side of the tube, and washed first with isotonic εaline containing 0.1% Triton X-100 and 0.1% bovine serum albumin, and then with Sau3a restriction enzyme digestion buffer. The DNA was then suspended in 20 μl Sau3a digestion buffer (New England Biolabε) and digeεted with 3U of Sau3a in order to releaεe the εubtelomeric DNA, leaving the terminal restriction fragments attached to the beads. The bead-TRF complexeε were removed with a magnet, and the supernatant containing the subtelomeric DNA was heated at 70°C for one hour to inactivate the Sau3a. PCR linkers were added to the subtelomeric DNA fragments by adjusting the buffer to 5mM DTT and 0.5 mM ATP, adding 25 pmol annealed PCR linkers pluε 1.5 U of T4 DNA ligaεe, and incubating overnight at 16°C. The sequence of the PCR linkerε used iε : OLM2: 5' TGGTACCGTCGAAAGCTTGACTG 3' DM01: 3' ATGAACTGACCTAG 5'

Theεe linkers are designed such that the annealed linkers have a Sau3a compatible end (5' GATC 3') , the 3' end of OLM2 will become ligated to the subtelomeric DNA fragment, while the 5' end of DM01 (which is not phoεphorylated) will remain unligated. The overlap between OLM2 and DM01 has an approximate melting point of 24°C, so that heating the ligated mixture to 70°C for 20 minuteε both inactivates the ligase and disεociateε DM01. Half of the ligation mix waε then diluted in PCR buffer with 100 pmol OLM2/l00μl aε the only primer. After three thermal cycleε of 72°C x 1 min then 85°C x 1 min (in order to fill in the complementary sequence to 0LM2 before melting the DNA) the DNA was PCR amplified for 20 cycleε (94°C x 1 min, 55°C x 1 min, 72°C x 3 min) .

The purity of the PCR amplified εubtelomeric library waε aεsessed by in situ hybridization to metaphaεe chromoεomeε. Three probeε were prepared by amplifying the librarieε in the presence of digoxigenin labelled UTP: a positive control in which PCR linkers had been ligated to a concatenated TTAGGG oligonucleotide to produce an amplified mixture containing an average size of about 1 kbp of telomeric repeats ("Concatenated GTR") ; a negative control of the

DNA selected with the biotinylated TTAGGGTTAGGG primer

("GTR-selected") ; and the experimental library selected with the biotinylated CCCTAACCCTAA primer

("CTR-selected") . The slides were hybridized to the different probes, stained with an anti-digoxigenin monoclonal antibody followed by an alkaline phoεphataεe conjugated anti-mouεe andibody, then coded and εcored for the preεence of εignal at internal εiteε versus telomeric ends. Only after being analyzed was the code broken. The reεults are shown in Table 7:

Table 7: In Situ Hybridization Analysis of Subtelomeric DNA (two experiments)

Probe End Signal Internal %Telomeric

Signal Concatenated 104, 46 20, 19 81%, 71% GTR GTR-selected 20, 32 90, 95 18%, 25%

CTR-εelected 76, 79 57, 29 57%, 73%

The CTR-εelected PCR amplification productε were then cloned, and 37 individual cloneε were picked and analyzed by in εitu hybridization. 10/37 (27%) of theεe cloneε gave telomeric εignalε. The reason why a much smaller fraction of the individual cloneε were telomeric than the fraction of signals in Table 7 iε due to the complexity of the PCR amplified material:

Actual telomeric DNA would be relatively abundant and thuε be able to give a εignal, while contaminating internal εequenceε would be highly diverεe and thuε each individual εequence in the mixture would tend to be too rare to give a εignal. The 20kbp of DNA at the end of each of 46 chromosome endε repreεents approximately 1/3000 of the genome. The telomeric location of approximately 1/3 of the cloned CTR-enriched DNA thus indicates that using the

' 5 biotinylated CTR resulted in a 1000-fold enrichment for telomeric DNA.

Seven of the telomeric clones were present on individual telomeres, while three hybridized to multiple telomeres . The characteristicε of the ten

10 telomeric clones are listed in Table 8, and partial DNA sequences from all but clone CSITU6 are shown in Table 9.

Table 8. Characteristics of Subtelomeric Clones

15 Clone Approx. Size Number of

Telomeric Signalε

CSITU5 1.5 Kbp εingle

CSITU6 0.5 Kbp multiple

CSITU9 0.9 Kbp εingle

CSITU13 0.9 Kbp single

20 CSITU22 0.9 Kbp multiple

CSITU24 0.9 Kbp multiple

CSITU33 0.8 Kbp single

CSITU37 0.9 Kbp single

CSITU38 0.9 Kbp single

25 CSITU51 1.5 Kbp single Table 9. Sequences of Subtelomeric 5a Clones

CSITU5 1 GATCTAGGCACAGCTGCTTCTCATTAGGCAGGTCTCAGCTAGAAGACCAC

51 TTCCCTCCCTGAGGAAGTCAACCCTTCTGCCACCCCATGGCCTTGCTTAAA

101

TTTTCAGACTGTCGAATTGGAATCCTACCTCCATTAGCTACTAGCTTGGG

151 CAAGATACAGAGCCCTCCC

Total number of bases is : 169

DNA sequence composition: 39 A; 54 C; 33 G; 43T;0 OTHER

CSITU9 1 ATATATGCGCTACATAAATGTATCTAGATGCAATTATCTAGATACATATA

51 AGAAAGTATTTGAAGGCCTTCTACAAGGCTTAGTTATTATATTGGTTCAT

101 ACAAGTTCTTCTTCAG

Total number of baεeε is: 116

DNA sequence composition: 39 A; 17 C; 18 G; 42 T; 0

Other

CSITU13

1 ATCCTTCTCCGCAAACTAAC-AGGAACAGAAAACCAAACACTGCATGTTCT

51 CACATCATTGTGGGAGTTGAACAATGAGAACACATGGACACAGGGAGGGG

101 AACATCACACACTCGGGGTGTCAGCCGGGTGGGAGGGTAGAGGAGGAGAA

151 ATACCTAAGTTCCAGATGACAGGTTG Total number of bases is : 176

DNA sequence composition: 58 A; 37 C; 50 G; 31 T;

Other

CSITU22

1 GATCTATGCTACCTCTAGGGATGGCACCATTCACAAGCACAAAGGAGATG

51 TCAGTGATTAAAAACACATGCTCTGGAGTCTGAGAGACTTTGAGACTTGC

101 TAGCTTGTGACTCTGCAGAGTTTAAGGTATCTGGACCCCTTTTTCCCTCA

151 TGTGCATAATGAAGAGATT Total number of baseε is: 169 DNA sequence compoεition: 47 A; 35 C; 39 G; 48 T; Other

CSITU24 1 GATCAACACTGTTAGTTGAGTACCCACATCACAAACGTGATTCTCAGAAT

51 GCCTTCCTTCCTGTCTAGTTTCTATAGGTAGATATTTCCTTTTTCAGCAT

101 AGGCCTGAAAAGCCGCCTCCAAATGCCCGCCTTCCAGACACTATAAAAAG

151 AGGGTTCAAACCTACTCTATGAAAGGGAATGTTCAACACAGA

Total number of baεeε is : 192

DNA sequence composition: 58 A; 49 C; 33G; 52 T; 0 Other

CSITU33

1 GATCTGTTTATTATTCTTCCAATATCTCCCCATCTCTTAAAAATTGGTTA

51 TTTCTTCGTTCATACATTTTTATCTCCCAAATTANNNNTGAGACTGGTTT

101 GAAGAGAGGAAAGCAATGTACACACTTTTATATTCCACCATGTATATCCG 151 GATATCC

Total number of baεes is: 157

DNA εequence compoεition: 43 A; 32 C; 19 G; 59 T; 4

Other

CSITU37

1 AATCCTCCTACCTTAACCTCCCTTTGTTAGCCTGCCATTACAGGTGTGAG

51 CCACCATTGCTCATTCGTCCGTTTATTCATTCAACAAATCAATCGATCTA

101 TTACATGTGAGGGACTCTTCAGGTCATGGGAATTC

Total number of baεeε iε: 135

DNA εequence compoεition: 32 A; 37 C; 22 G; 44 T;

Other

CSITU38

1 GATCACTTGAGCCCAGGAGTTTGAGACCAGCCTGGGTGACATGGCAAAAC

51 CCCATCTCTACCAAAAGAAAAAAANNNNACAAATTGGTGGTGTTGATGGT

101 CGGCGACCATTGATCCC

Total number of baεeε iε : 117

DNA sequence composition: 35 A; 27 C; 28 G; 23 T; 4 Other

CSITU51

1 GATCAGGGAGGGGCCGAAAACTGGAGATGCAGGTGTGCTGTAAGACACTG

51 CAGAGAGGGCATTTACCTGCCCCATCATCCAGCACAGGAACAGCGACTGA 101 CAGCGCTCACCCACCCACCATCGCCAGTCACACTGGG

Total number of bases is: 137

DNA sequence composition: 37 A; 42 C; 39 G; 19 T; 0 Other

The CTR-enriched subtelomeric PCR amplified library has also been uεed to screen a cDNA library. 32 clones have been isolated, and partial sequence haε been obtained form five cloneε. Their sequences are shown in Table 10.

Two of these clones, PhC4 and PHC5, have been characterized on Northern blots. Both hybridize to the same two mRNAs of approximately 6.2 and 7.7 Kb. Since the 3' εequences of PHC4 and PHC5 are different, thiε suggestε they may repreεent alternative splicing products of the same gene. Both mesεageε are abundant in PDL 38 IMR90 cells, which have relatively long telomeres, and neither is expressed in the immortal IDH4 cells (which have very short telomeres) that were derived from IMR90. This supports the hypothesis that the expression of genes located in the subtelomeric DNA are regulated by telomeric length. Thiε data is evidence that the above mentioned procedure provides a means of obtaining sequences located in the proximity of telomeres, some of which encode mRNA. Those εequenceε which are unique to individual chromosomes will be useful in genomic mapping. Those which are active genes and differentially expressed in cells with differing telomere length, may play an important role in communicating information relating to telomere length to the cell. Genes that regulate the onset of Ml senescence can be isolated by theεe meanε, aε will aε genes which modulate telomerase activity. The function of the telomeric geneε can be identified by overexpreεεion and knock-out in young senescent and immortal cells. Such cDNAε, antiεenεe molecules, and the encoded proteins may have important therapeutic and diagnostic value in regard to their modulation of cell proliferation in age-related disease and hyperplasias such as cancer.

Table 10. Partial Sequence of subtelomeric cDNA clones .

PHC4-5'end

1 GGCTCGAGAACGGGAGGAGGGGGCTCTTGTATCAGGGCCCGTTGTCACAT

51 CTGCTCTCAGCTTGTTGAAAACTCATAATC

Total number of bases is: 80

DNA εequence composition: 17 A; 19 C; 24 G; 20 T; 0

Other

PHC5-3ΕND

1 AGGTCCCTTGGTCGTGATCCGGGAAGGGGCCTGACGTTGCGGGAGATCGA

51 GTTTTCTGTGGGCTTGGGGAACCTCTCACGTTGCTGTGTCCTGGTGAGCA

101 GCCCGGACCAATAAACCTGCTTTTCTTAAAAGGAAAAAAAAAAAAAAAAAA

151 AAAAAAA

Total number of bases is: 157

DNA sequence composition: 47 A; 31 C; 44 G; 35 T; Other

PHC7

1 ATCTAGGTTTTTTAAAAAAGCTTTGAGAGGTAATTACTTGCATATGAGAG

51 AATAAAACATTTGGCACATTGTTAAAAAAAAAAAAAAAAAAAAAAAAAAA

101 AAAAAAAAAAAAAAAAAAAA

Total number of bases is: 120 DNA sequence composition: 73 A; 7 C; 14 G; 26 T; Other

PHC8

1 CTCATTTACTTTTCTCTTATAGCGTGGCTTTAAACATATATACATTTGTA

51 TATATGTATATATGAATATAATGTATAAAATGTATGTAGATGTATATACA

101 AAAAATAAACGAGATGGGTTAAAGATATGTAAAAAAAAAAAAAAAAAAA

Total number of bases is: 149

DNA sequence composition: 69 A; 11 C; 19 G; 50 T; C

Other

PHC9

1 AGTCCCAGCTACTCGGGAGGGCTGAGGCAGGAGAATGGCGTGAACCCAGG

51 AGGCGAAGCTTGCAGTGAGCTGAGATCGCGCCACTGCACTCCAGCCTGGA

101 CGACAGAGCGAGACTCTGTCTCAAAAAAAAAAAAAAAAAAAA

Total number of baseε iε : 169

DNA εequence compoεition: 47 A; 35 C; 39 G; 48T; 0 Other

Example 19 : Iεolation of Factorε that Derepress Telomeraεe

The M2 mechanism of cellular senescence occurs when insufficient numbers of telomeric repeats remain to εupport continued cellular proliferation. Eεcape from the M2 mechanism and immortalization occur concomitantly with the induction of telomerase activity and stabilization of telomere length, and thuε the inactivation of the M2 mechaniεm directly or indirectly derepresses telomeraεe. The gene(s) regulating the M2 mechanism have been tagged with retroviral sequenceε . The methods by which this was accomplished conεiεted of first determining the frequency at which a clone of SV40 T-antigen transfected human lung fibroblastε was able to escape M2 and become immortal (T-antigen blocks the Ml mechanism, thus the M2 mechanism is the εole remaining block to immortality in theεe cellε) . The pre-criεiε cellε were then infected with a defective retroviruε in order to inεertionally mutagenize potential M2 geneε, and it waε shown that the frequency of immortalization was increased by almoεt three-fold. Finally, pulεe-field electrophoreεiε of different immortalized inεertionally mutagenized lineε was used to identify which of the lines became immortal due to an insertion into the same M2 gene. Since an M2 mechanism gene has now been tagged with retroviral sequenceε, thoεe with ordinary skills in the art can now clone and identify the specific gene. The methodε uεed were as follows:

The frequency of escape from crisiε (e.g., the immortalization frequency of T-antigen expreεεing cells) was estimated uεing an approach baεed on what iε eεεentially a fluctuation analyεis aε previouεly deεcribed (Shay, J.W., and Wright, W.E. (1989) Exp. Cell Reε. 184, 109-118) . SW26 cellε (a clone iεolated from IMR 90 normal human lung fibroblaεtε tranεfected with a vector expreεεing SV40 large T antigen) were expanded approximately 15 PDL'ε before criεiε into multiple series at a constant cell density of 6667 cells/cm². Each εerieε waε εubεequently maintained aε a εeparate culture, εo that at the end of the experiment the fraction of each εerieε that gave riεe to immortal cell lineε could be determined. Cultureε were split at or just prior to confluence at 6667 cellε/cm². Once cellε reached criεiε they were εplit at leaεt once every three weeks until virtually no εurviving cellε remained or the culture had immortalized. When too few cells were obtained, all of the cells were put back into culture in a single dish. Fibroblaεts were considered immortal if vigorous growth occurred after criεiε during two εubcultivations in which 1000 cells were seeded into 50 cm² diεheε and allowed to proliferate for three weekε for each cycle.

SW 26 cellε enter criεis at approximately PDL 82-85. Numerous vials of SW26 cellε (8 x IO⁶ cells/vial) were frozen at PDL 71, and teεting verified that spontaneouε immortalization eventε had not yet occurred. Five vials were thawed, scaled up for 4 days to approximately IO⁸ cells (thus to approximately PDL 74) , then trypεinized and combined into a εingle pool of cellε in 40 ml of medium and diεtributed into 200 10cm² dishes. Thirty dishes were treated with 25μg/ml bleomycin sulfate (a chemical mutagen) for two hours in serum free medium one day later. Since this concentration of bleomycin sulfate resulted in approximately 50% of the IMR-90 SW26 cells dying, theεe diεheε had been plated at twice the cell denεity as the rest.

The remainder of the dishes were used as controls (70 disheε) or infected with LNL6 defective retrovirus (100 dishes) . LML6 was generated in the amphotrophic packaging line PA317 according to previously described procedures (Miller and Rosman, 1989, Biotechniques 1_, 980-990) . Culture supernatant from LNL6 infected PA317 cells were used to infect one hundred disheε containing approximately 5 x 10⁵ cellε in the preεence of 2μg/ml of polybrene. Control medium supernatant from uninfected PA317 cells containing polybrene were used to treat 70 dishes and served aε controls. Within a few days after infection, all control and experimental dishes were counted and each dish contained 1-2 x 10⁶ cells. The PDL of each dish was calculated and cellε were then replated at 0.33 x IO⁶ in 50 cm² diεhes and maintained separately to conduct the fluctuation analyεiε.

Bleomycin treated SW26 cellε eεcaped crisis with an approximately two-fold higher frequency (7.7 x IO^"7) than the spontaneouε rate (4.7 x IO^"7) . Pre-criεiε SW26 cellε infected with the defective retrovirus LNL6 in order to produce insertional mutationε yielded a frequency of eεcape from criεiε (10.9 x IO^"7) that waε 2-3 fold greater than the rate from simultaneous control series mock-infected with culture supernatant from the non-infected packaging line. Table 10 Bleomycin Sulfate Exposure and Retroviruε Infection Increase Immortalization Frequency Addition Immortalization Frequency

Nil 10/68 4.4 x 10

Bleomycin 7/27 7.7 x 10 -7 sulfate

LNL6 retroviruε 36/99 10.9 x IO^"7 Immortalization is expresεed as the number of immortal lines per number of culture series, each series being derived from a εingle diεh at the initiation of the experiment. Frequency iε expreεsed as the probability of obtaining an immortal cell line based on the number of cells plated at each paεsage (not per cell division) .

DNA has been isolated from 23 of the 36 independent cell lines obtained following insertional mutageneεiε with LNL6, and 7 of theεe (30%) did not contain retroviral εequences when analyzed on Southern Blots, while moεt of the remainder contained εingle insertions. Given that those without retroviral insertionε had to represent spontaneous immortalization events, most of the remaining clones with retroviral inεertionε εhould be due to inεertional mutageneεiε if the frequency of immortalization waε actually increaεed 2-3 fold. DNA from 12 lineε haε been digested with the rare-cutting enzyme Sfil, followed by pulse-field electrophoreεis, transfer to nylon membraneε and probing with the retrovirus LTR. Six of the 12 lines contained a common band of approximately 350 Kbp that hybridized to the retroviral LTR. Four of these six have alεo been analyzed following BamHI digeεtion, and three of theεe four also contained a common band of approximately 20Kbp. Given that the retrovirus is 6Kbp long, this εtrongly εuggests that the retrovirus haε inserted multiple times within 14 Kbp region of DMNA, which is strongly suggeεtive of a εingle gene. Digestion with EcoRl, which cuts within the retrovirus, yields different size fragmentε for each line, eεtabliεhing that they represent different insertional eventε and are truly independent isolates. The use of retroviral sequences to clone the genomic DNA flanking the insertion sites should now permit positive identification of a gene involved in the M2 mechanism. Interference with the function of that gene (for example, using antisenεe techniqueε) εhould reεult in the derepression of telomerase and the ability to extend the lifespan of normal human cells. This gene should also prove to be mutated in a variety of cancer cells, and is thus likely to be of diagnostic and therapeutic value in cancer as well.

Example 20: Tissue Distribution of Telomerase Activity in Primates

S100 extracts were prepared from a 12 year old healthy male Rhesuε Macaque to determine the tiεsue diεtribution of telomerase activity. Abundant telomerase activity was detected only from the testiε. Sampleε of tiεsue from the brain, kidney, and liver displayed no detectable activity. This suggests that telomerase inhibition as a therapeutic modality for cancer has the unique advantage of not being abundant in normal tisεueε with the exception of the germ line. Therefore telomerase inhibitors should be targeted away from the germ cells in reproductive aged individuals to decrease the chance of birth defects. Such targeting may be accomplished by localized injection or release of the active agent near the site of the tumor. The effect of the telomerase inhibitors in the male may be easily determined by measuring telomere repeat length in the sperm.

Example 21: PCR Assay for Telomeraεe Activity

In normal εomatic cells other than germ-line and some εtem cellε telomeraεe activity is not detected and with each cell division the chromosomes lose 50 to 200 nucleotideε of telomeric εequence, consistent with predictions of the end replication problem. Eventually all normal cells cease to divide, and this state of replicative senescence is thought to be triggered by a critically short telomere length. In germ cells and immortal cells telomerase is active, telomere length is maintained, and replicative senescence does not occur. It is hypothesized that by controlling telomerase activity, telomere length could be modulated to ultimately impact the processes of cellular senescence and immortalization.

In vitro studies of telomerase rely on the enzymes ability to template and catalyze the synthesis of telomeric sequence onto a single-stranded oligodeoxynucleotide (oligo) substrate. The conventional assay for this activity uses an oligo of known sequence as substrate, radioactive deoxynucleotide triphosphate (dNTP) for labeling, and sequencing gel for resolution and display of the products . Since telomerase stalls and can release the DNA after adding the first G in the T₂AG₃ repeat, the characteristic pattern of products is a six nucleotide ladder of extended oligo substrate. The phase of the repeats depends on the 3' end sequence of the oligo substrate; telomerase recognizes where it is in the repeat and synthesizes accordingly to yield contiguous repeat sequence. Although telomeric sequence oligos are the most efficient in vitro substrates, telomerase will also synthesize T₂AG₃ repeats onto non-telomeric oligos.

Originally developed for the Tetrahymena enzyme and then adapted for the human and mouse enzymes, this assay is highly specific, easily controlled, detects non-processive and processive activity, and continues to provide useful biochemical and enzy ological information on telomerase. Despite its utility, the conventional assay has several drawbacks including insensitivity, radioactivity, labor and time intensiveness, and the need for specialized equipment and expertise. In the standard εize reaction a quantity of immortal cell extract equivalent to 10^s cellε minimum is required for unambiguous detection of activity. Using relatively high levels of radioactivity (30 μCi) , the quantity of labeled product is sufficiently low to require several dayε film expoεure for autoradiography or expensive phosphorimager technology for overnight results. An experienced bench scientiεt spends most of a working day completing 20 to 40 assay reactions with only one significant free block of time (during the gel run) .

With the goal of retaining the strengths while improving on the drawbacks of the conventional aεεay, we developed a novel assay for telomerase activity. The basis of the new asεay iε application of the polymeraεe chain reaction (PCR) for εpecific amplification of the in vitro products of telomerase. The result is a simple and rapid procedure at least 100 times more sensitive than the conventional assay with a detection limit of IO⁶ molecules of telomerase productε or 1000 telomerase positive cells. Several drawbacks of the conventional telomerase assay collectively could be improved by increasing the senεitivity of the aεεay. Since the productε of telomerase are composed of a specific sequence of DNA, a conceptually simple approach to sensitizing the in vi tro aεsay is PCR amplification of the telomerase products. The following scheme was evaluated.

The telomerase reaction portion of the assay is unchanged. The telomeraεe products- the nested set of one to hundreds of εix nucleotide telomeric repeatε added to the oligo εubεtrate- εerve as the templates for PCR amplification. Exponential amplification is achieved by a PCR primer set consisting of a downstream primer complementary to telomeric repeatε and an upstream primer composed of sequence from the oligo substrate. In fact, the oligo substrate (typically an 18-mer) itself serves as the upstream primer and since a standard telomeraεe reaction containε IμM oligo substrate of which less than 1% is extended by telomerase, there is no need to add more for the PCR reaction. A downstream primer of comparable length is used. A stringent annealing temperature in the PCR cycle ensures specific primer binding which results in preservation of the six nucleotide ladder in the PCR productε, reflective of the template population. For at least three reasonε, the PCR products are not directly proportional to the telomerase products: (i) the downstream primer (e.g. 18-mer) can anneal perfectly at more than one position on discreet telomerase products longer than three repeats; (ii) the template population iε a diεtribution of lengths, concentrations, and templating efficiencies,- and (iii) the PCR favors syntheεiε of εhorter products.

The oligo substrates first analyzed for use in a PCR-based asεay were telomeric repeat sequenceε, the moεt efficient in vitro subεtrateε for telomerase (e.g. (T₂AG₃)₃) . Since the downstream primer (e.g. (C₃TA₂)₃) iε complementary to telomeric repeatε it will anneal not only to telomeraεe products as desired but also to the unextended oligo substrate. In theory if conditions were sufficiently stringent to allow only perfect annealing, then the oligos alone could only form a perfect duplex and would not be extended by Taq polymerase; a telomerase product would be neceεεary to provide a duplex with a receεεed 3' end for extenεion by Taq polymerase. In practice such conditionε were not found.

In PCR reactionε containing (T₂AG₃)₃ and (C₃TA₂)₃ alone, various conditions tested for perfect annealing alεo allowed 2 of 3 telomeric repeatε to anneal in a εtaggered alignment providing the εubstrate for Taq polymerase. The PCR products of thiε firεt cycle are (T₂AG₃)₄ and (C₃TA₂)₄. Staggered annealing in εubεequent cycles leads to a six nucleotide ladder of products that extends hundredε of nucleotideε in length (data not εhown) . This "false positive" result would be indistinguishable from PCR amplification of telomerase products hence a different approach was necessary. Recognizing that telomerase also εyntheεizeε T₂AG₃ repeats onto non-telomeric oligos, we employed such oligo εubεtrateε to avoid PCR primer complementarity.

The sequences of three oligo substrates and their first four products which reεult from in vitro extension by telomerase are shown in Fig. 34. (T₂AG₃)₃ and (GT₂AG₂)₃ are typical oligo substrateε used in the conventional assay. Comparing the first productε of theεe telomeric εubstrates illuεtrateε the ability of telomeraεe to recognize itε position within telomeric repeat sequence (Fig. 34, 1st products) . In a conventional assay of a detergent extract from immortal 293 cells, the six nucleotide ladder of products from (T₂AG₃)₃ was phased one nucleotide shorter than that from (GT₂AG₂)₃ (Fig. 35, lanes 1,2). Generation of products was sensitive to pretreatment of the extract with RNase (laneε 3,5). The εix nucleotide ladder, the dependence of the product phaεe on the telomeric substrate, and the sensitivity to RNase pretreatment identified this activity as telomerase. The oligo substrate M2 contains a five of six identity with telomeric sequence at its 3'end but no other telomeric sequence (Fig. 34) . As expected from previous work, this non-telomeric oligo served as an efficient in vitro substrate for telomerase (Fig. 35, laneε 4,5) . The εequence of telomerase products of the M2 oligo substrate (as shown in Fig. 34) was confirmed by chain termination sequencing (data not shown) .

The εecond role of oligo M2 in the PCR-baεed aεεay iε to εerve aε the upεtream PCR primer. When paired with an appropriate downstream primer (complement of telomeric repeats) specific amplification of telomerase productε of M2 would reεult. Moεt importantly, M2 abεolutely must not anneal with the downstream primer. Thiε iε becauεe even minor levelε of primer annealing can yield firεt cycle PCR productε identical to telomerase products (i.e., M2 plus (T₂AG₃)_n) . In subsequent cycleε theεe productε would template the production of a εix nucleotide ladder of PCR products resulting in a false positive. As described above, this problem was first countered by choosing a non-telomeric oligo subεtrate/upεtream primer. However, further meaεureε were necessary to fully quench annealing of the primers.

Several rounds of oligo design and experimental trialε resulted in the downstream primer designated CX (Fig. 34) . CX is composed of sequence complementary to three imperfect telomeric repeats and one perfect repeat. The single nucleotide difference in three of the repeats compromises the capacity of CX to anneal to the 3' end of M2 (which contains 5 of 6 nucleotides of a telomeric repeat) . Moreover, any poεεible alignment between theεe primers nucleated by the telomeric complementarity leadε to a duplex in which the recessed 3' nucleotide is mismatched. To further discourage primer interaction the T4g32 single-stranded binding protein, known to diminish primer dimer formation, was included in the PCR reactionε. For variouε primer εetε we obεerved a three- to five-fold decrease in primer interaction when T4g32 protein was added. Under these conditions, M2 and CX alone in a PCR reaction set up at room temperature and then subjected to 27 cycles of 95", 50°, and 72° produced no PCR productε (Fig. 36, Lane 3) . As yet another measure to prevent primer interaction and non-specific amplification, the hot start method was utilized. In our adaptation of this technique, CX was dried at the bottom of the tube and then covered with a wax barrier. All other PCR reaction components were combined in the tube above the wax barrier, and the tube was placed in the thermal cycler. With this εet-up, CX did not appear in the PCR reaction until the wax melted (about 60°C) during the firεt cycle, preventing CX interaction with any other reaction component at a temperature below the annealing temperature. As expected, this additional precaution combined with the described conditions yielded no PCR products from primerε alone (lane 4) . To teεt whether theεe conditionε would allow the specific amplification desired, synthetic oligoε representing the first four telomerase products of M2 were obtained. The sequenceε are εhown in Fig. 34 and the oligoε were deεignated M2+1, M2+2, M2+3, M2+4. Under these conditions a three telomeric repeat extension of M2 was the minimal requirement for amplification (Fig. 36, lanes 8 - 11) . The PCR products from amplification of M2+3 (lane 9) and M2+4 (lane 10) were six nucleotide ladders extending from 40 nucleotides up to the limit of gel resolution. The 40 nucleotide product resulted from alignment of CX and M2+4 as shown in Fig. 34. The very same alignment of CX and M2+3, held together by three repeats, generated the 40 nucleotide PCR product. Since these conditions allowed annealing by 3 of 4 repeats, staggered annealing also occurred which led to generation of the six nucleotide ladder. If a telomerase product is of sufficient length for primer annealing under the chosen conditionε, then amplification occurs. The ladder of PCR products means only that thiε criterion haε been met, and doeε not provide information on the ladder of telomeraεe products.

Having modeled the amplification conditions with synthetic telomerase products, we next tested authentic telomeraεe products. According to the conventional procedure, telomerase aεεay reactionε of an immortal cell extract were carried out uεing M2 as the oligo subεtrate. After the reactions were fully proceεεed in preparation for εequencing gel analysis, 1/10 of the purified telomerase products was removed and subjected to PCR amplification. The rest waε loaded on a εequencing gel to complete the conventional aεεay. The results of the conventional asεay are εhown in Fig. 36, laneε 1 and 2. M2 oligo substrate was efficiently extended by telomerase yielding a εix nucleotide ladder of productε (lane 2) and the activity waε εensitive to RNase pretreatment of the extract (lane 1) . Using the PCR conditions described above, 1/10 of the products of the RNase pretreated reaction produced no PCR productε (lane 5) . 1/10 of the telomeraεe positive reaction products subjected to PCR conditions without the downstream primer yielded no detectable signal (lane 6) . When the primer waε provided, PCR amplification of authentic telomerase products occurred (lane 7) and waε indistinguiεhable from PCR amplified synthetic telomerase products.

In the conventional aεsay procedure a 40 μl telomerase reaction is set up and incubated for 60 - 90 minuteε, the reaction iε terminated, and then εeveral proceεεing εtepε are carried out to purify telomeraεe productε for εequencing gel analyεiε. PCR amplification of the telomeraεe productε at this final stage was highly efficient. In order to reduce the time and number of manipulations in the assay, we tested whether telomerase products in a leεε purified and concentrated εtate would serve aε efficient templates in PCR amplification. Immediately after the incubation period of the telomeraεe reaction, a 2 μl aliquot waε removed and εubjected to PCR amplification. Thiε reεulted in εpecific amplification of the unpurified telomeraεe productε that was indistinguishable from PCR of purified productε (data not shown) .

Since both telomerase and Taq polymerase are DNA synthesizing enzymes with similar reaction components, the asεay could be further streamlined by combining the activities in a single reaction. A single tube protocol was achieved and is shown schematically in Fig. 37. The CX oligo is iεolated by wax barrier for hot εtart of the PCR. All other reaction componentε are combined above the wax barrier including the telomeraεe oligo substrate/upεtream primer M2, the telomerase extract, and Taq polymerase. PCR buffer and deoxynucleotide conditions allow sufficient telomeraεe product generation in 10 minuteε at room temperature. The tubeε are then εimply placed in the thermal cycler for PCR. As deεcribed above, specific amplification of telomerase products under these conditions occurε if and only if the oligo subεtrate M2 has been extended with three or more T₂AG₃ repeats.

Resultε from application of the single tube protocol are shown in Fig. 38, lanes 5 - 13. We first demonεtrated that in a conventional telomeraεe aεsay the M2 oligo was an efficient telomerase substrate when assayed in PCR conditions (lanes 1 - 4) . Using the single tube protocol, primers alone (lane 6) and immortal 293 cell extract alone (lane 5) gave no signal. The 293 extract assayed in the presence of the oligo primers produced the specific amplification products (lane 7) . When the 293 extract was pretreated in various ways known to inactivate telomerase including 65° for 10 minutes (lane 8), RNase (lane 9) , phenol extraction (lane 10) , and protease (lane 11) , no asεay signal was produced. An extract made from BJ cells ( a normal fibroblaεt cell εtrain) produced no εignal (lane 12) . Partially purified telomerase from DEAE chromatography of a 293 cell extract gave a positive signal (lane 13) . These and other results demonstrate that detection of telomerase by the PCR-based assay is entirely consistent with that by the conventional assay.

The relative senεitivity of the PCR-baεed telomeraεe aεεay and the conventional aεsay was compared. Serial dilutions of a 293 extract were teεted in both aεεays (Fig. 39) . With the conventional aεsay, the minimum quantity of extract necessary for telomerase product detection uεing optimal conditions was 0.2μl of the 293 extract (lane 4) . With the PCR-based assay a minimum of O.Olμl gave a positive signal (lane 8) . This corresponds to a 20-fold higher senεitivity in the PCR-baεed aεεay. However, there are additional factorε to take into account. Firεt, the expoεure time of the gel for the conventional aεsay waε at leaεt five timeε that for the PCR-based asεay (6 hr vs. 1 hr) . Second, the amount of radioactive dGTP used for product labeling in the conventional assay was 10 times that of the PCR-based assay (30 μCi vs. 3 μCi) . Third, all of the aεεay productε were loaded onto the gel in the conventional assay, where only half of the asεay productε were loaded onto the gel in the PCR-baεed aεεay. Conεidering theεe factorε, a conεervative estimate is that the PCR-based asεay iε at leaεt 100-fold more εenεitive.

The limit of εenεitivity of the PCR-baεed assay was analyzed by titration of the synthetic telomerase product M2+4, and titration of extracts from different numbers of 293 cells (Fig. 40) . Dilution series of M2+4 oligo was mixed with heat-treated (telomerase inactivated) 293 extract and analyzed in the PCR-baεed aεεay (Fig. 40, laneε 1-5) . In this analysiε, the PCR asεay gave a clear poεitive εignal from IO⁶ moleculeε of M2+4 (lane 2) . For evaluation of the extraction efficiency and the limit of detection from different cell numberε, extractε were made from a dilution εerieε of 293 cells. In the extractions, the amount of extraction buffer was kept constant (lOOμl) , while the total number of 293 cells was varied. These extracts were then tested in the PCR-based asεay (laneε 11-15) . The reεultε εhowed that the PCR-baεed assay can detect telomerase activity from as few as 10³ 293 cells (lane 12) . Furthermore, the result shows that a telomerase positive extract can be made by the detergent lysis method from as few as IO⁵ cells. This is an important application of the assay since very often the moεt interesting cells to teεt are available in only limited quantitieε. There waε no activity detected in the extract to teεt from normal fibroblaεt cellε (BJ, lane 10) even when ample quantitieε of the extract were tested. If present, telomerase activity in these cells is at least 1000 timeε lower than in 293 cellε. Dilutionε of the extract from 10⁶293 cells were also tested by the PCR-baεed aεεay (laneε 6-9) . By thiε titration, a εimilar limit of εenεitivity of IO³ cellε waε observed (lane 7) . Since the conventional asεay has a detection limit of IO⁵ 293 cellε, the 10³ cell limit by the PCR-baεed assay correspondε to our estimation of about 100-fold higher senεitivity of the PCR-baεed aεεay. A rough correlation can be drawn from the limitε of detection of at least IO⁶ molecules of synthetic telomerase product and at leaεt IO³ 293 cellε. Telomeraεe activity extracted from each 293 cell extended a minimum of IO³ molecules of M2 oligo with at least three telomeric repeats in ten minutes.

The PCR-based telomerase activity assay described here provides several significant advantages over the conventional assay. First, it is several orders of magnitude more sensitive. Second, the reactions are less labor intensive and faster to complete. Third, the results are more readily obtained. Fourth, little or no radioactivity is required. And finally, the methodology lends itself to further significant improvements, including a single-cell assay for telomerase activity in vivo .

One familiar with the art can readily modify the current technology such that false positives which may occur when incorrect reaction conditions are uεed will be detected. For example, input oligonucleotideε can be engineered εuch that primer-dimer productε and the PCR ladder they can generate under sub-optimal conditions do not align with a telomerase-generated PCR ladder. It is also possible to create a quantitative assay such that the telomeraεe-dependent PCR productε are proportional to the amount of initial telomeraεe activity, and to increaεe the εenεitivity such that activity in a single cell could be detected. Since nano-microgram amounts of double-stranded DNA can be generated, one can readily use non-radioactive detection systems, εuch aε fluoreεcence, thuε alεo providing the opportunity to create a "single-tube" assay.

Single-cell assays could be done with the methods deεcribed above in which a cell-free extract iε generated prior to primer extension. However, it is also possible to incubate viable cells with the subεtrate oligonucleotide following which the oligonucleotide will be extended if the cell poεεeεεes functional telomerase-activity. Eεtablished in εi tu PCR technology with Taq polymeraεe, the C-rich PCR primer, and labeled precurεors could then be used on fixed cells to amplify telomeraεe-extended substrate oligonucleotides. Telomeraεe poεitive cellε would be detected by microscopy utilizing incorporation of the labeled nucleotide during PCR amplification.

The major applications for the PCR-based telomerase aεεay are in reεearch and diagnostics. Since the aεsay is fast, simple, and amenable to εingle-tube reactions and in εi tu detection, it can be used in research and clinical laboratory settingε where there iε need to detect telomeraεe poεitive cellε. Such applicationε include, but are not limited to: (i) Detection of immortal cellε in tumor biopεieε for the identification of potential metaεtatic cellε. (ii) Identification, in a cell baεed εcreen, of agents capable of derepressing telomeraεe. Such agents include immortalizing agentε ( e . g. oncogeneε) or compoundε which might be selected for their ability to transiently activate telomerase and hence extend telomeres and replicative lifespan of cells. (iii) Identification in culture systemε, or in vivo, of stem cells or early progenitor cells which possesε telomerase activity, (iv) Examination of telomerase regulation during differentiation and development, (v) Identification of telomerase-poεitive fractionε generated during purification of telomeraεe. (vi) Identification of protozoal or fungal infectionε through the use of specific primers to detect the preεence of telomeraεe-positive eukaryotic pathogens, (vii) If human sperm cellε are telomeraεe poεitive, it may be posεible to diagnoεe certain typeε of infertility characterized by a failure to activate telomeraεe during gametogeneεiε.

Exampleε of εome of theεe applicationε include our detection of telomeraεe in CD34⁺ hematopoietic stem cells, and the detection of a weak telomeraεe activity in total peripheral blood leukocyteε which apparently reflectε the circulating population of theεe cellε in blood. Prior to our uεe of the PCR telomeraεe aεεay, telomeraεe activity had never been reported in any non-transformed or non-germline cell type. A second example includes the use of the PCR-based telomerase aεεay for following activity in column chromatography during purification of telomeraεe-poεitive extracts.

The following materialε and methodε were uεed in thiε example: Materials

PCR-based assayε were performed in 0.2ml Strip-ease tubes from Robbins Scientific (Sunnyvale, CA) which were autoclaved before use. All oligodeoxynucleotides were Ultrapure grade (HPLC purified) obtained from Keystone Laboratory (Menlo Park, CA) which were suεpended in H₂0 at a concentration of lmg/ml. Taq DNA polymeraεe, Tween 20, and T4g32 protein were from Boeringer Mannheim. Radioiεotopeε were from NEN-Dupont. dNTPε from Pharmacia were aliquoted, εtored at -20°C, and thawed no more than twice before use. All other reaction components were molecular biology grade from Sigma except when otherwise noted. Diethylpyrocarbonate-treated, de ionized, sterile H₂0 was used throughout the experiments.

Extract preparation

Cells used in thiε εtudy were 293 cellε, an immortalized line derived from human embryonic kidney cellε tranεformed with fragments of adenovirus type 5 DNA; and BJ cellε, a normal cell εtrain of human εkin fibroblaεts. Cells were grown in Joklik's medium containing 5% (293) or 10% (BJ) fetal bovine serum. Adherent cell cultureε were grown to 80% confluency, harveεted, and extracted by the

CHAPS (3-{ (3-Cholamidopropyl)dimethylammonio}-l-propanesulfonat e, from Pierce) detergent lysis method (Ho and Prowse, unpubliεhed data) . A maximum of 1X10⁶ cellε were waεhed once in PBS, pelleted at 10 OOOg for 1 min at 4°C, and resuspended in 1 ml of ice-cold wash buffer [lOmM HEPES-KOH (pH 7.5), 1.5mM MgCl₂, lOmM KC1, lmM DTT] . The cells were pelleted again and reεuεpended in ice-cold lysis buffer [lOmM Tris-HCl (pH 7.5), ImM MgCl₂, ImM EGTA, O.lmM PMSF, 5mM /3-mercaptoethanol, 0.5% CHAPS, 10% glycerol] at a concentration of 20μl of lysis buffer per 1X10⁴ cellε. The suεpenεion was incubated on ice for 30 min. and then spun in a microcentrifuge at 10 OOOg for 20 min. at 4°C. The supernatant was removed to another tube, quick-frozen on dry ice, and stored at -70°C. These extracts typically contained 5 to 10 mg/ml total protein concentration. Telomerase activity was stable to multiple freeze-thaws. For the experimentε εhown in Fig. 40, extractε were made from different numberε of cellε. In these extractionε the lyεis volume of lOOμl was kept conεtant with different cell numbers.

Extract Pretreatments

Extracts were treated in variouε ways to inactivate telomerase. Heat treatment waε 10 min. at 65°C. RNaεe treatment was incubation of lOμl extract with 0.5 μg RNase (DNase-free, Boeringer Mannheim) for 10 min. at room temperature. For phenol treatment, the extract waε vortexed with an equal volume of phenol:chloroform (1:1), centrifuged, and the reεulting aqueous phase was used for analysiε. Protease treatment waε incubation of 50μl extract with 5μg Bromelain proteaεe (Boeringer Mannheim) for 10 min. at 37°C. Afterwardε, the Bromelain proteaεe in the extract waε removed by incubation with carrier-fixed α.₂-mackroglobulin (50μl of εettled gel correεponding to 1.25 mg. protein, Boeringer Mannheim) for 30 min. at room temperature with shaking. Then the c-₂-macroglobulin/Bromelain complex was pelleted by centrifugation for 10 min. at 10 OOOg, and the resulting supernatant was uεed for analyεiε. Conventional telomerase assay

The procedure and conditions of the conventional telomerase assay were as deεcribed by Morin (59 Cell 521- 529, 1989) . Oligo εubεtrateε were added to a concentration of lμM.

Preparation of wax-barrier reaction tubes

For hot-εtart PCR, reactionε were performed in tubes which contained lyophilized CX primer (5'-CCCTTACCCTTACCCTTACCCTAA-3') separated from the rest of the reaction components by a wax barrier. Tubes were prepared by adding 2μl of a 50ng/μl suεpenεion of CX primer (0.1 μg) which waε εpun to the bottom of the tube and evaporated until dry in a Speed-Vac. Tubeε were then heated at 70°C, and 7-10 μl of molten wax (Ampliwax, Perkin-Elmerε) waε pipetted into the bottom of the tube. After the wax waε allowed to solidify at room temperature, the tubes were stored at 4°C. Tubes were warmed to room temperature before use. No effect on assay performance was observed using prepared tubes stored at 4°C for up to two months.

PCR amplification of telomerase products

50 μl reactions set up at room temperature in the prepared tubes contained 20mM Tris-HCl (pH 8.3), 1.5mM MgCl₂, 68mM KCl, 0.05% Tween 20, ImM EGTA, 50μM dNTPs, 344nM of M2 oligo (17.2pmol, 5' -AATCCGTCGAGCAGAGTT-3' ) , 0.5μM T4g32protein, and 2 U of Taq DNA polymeraεe. For radiolabeling of productε, 0.2-0.4μl of lOμCi/μl ³²P-dGTP and/or ³²P-dCTP (800 or 3000 Ci/mmol) waε added to the reaction. Then the tubeε were tranεferred to the thermal cycler (96 well Singleblock εyεtem, Ericomp) for 27 roundε of 94°C for 30 sec, 50°C for 30 sec, and 72°C for 1.5 min. One half of the reaction was analyzed by electrophoresiε in 0.5X TBE, 15% polyacrylamide non-denaturing gels. Visualization of the products was by ethidium bromide staining, autoradiography, or phosphorimaging (Molecular Dynamics) .

Templates (telomerase products) for amplification were added just before thermal cycling. Templates were synthetic telomerase productε (M2+1, M2+2, M2+3, or M2+4, O.lfmol per reaction), purified telomeraεe productε ( 1/10 productε from a 40μl conventional aεsay) , or unpurified telomeraεe productε (2μl from a 40μl conventional aεεay) . For a single-tube asεay, lμl (~10μg total protein) of cell extract was added to the reaction mix, and the reaction was incubated at room temperature for 10 min. before PCR amplification.

Compoεitions

Compositionε or productε according to the invention may conveniently be provided in the form of εolutions suitable for parenteral or nasal or oral adminis¬ tration. In many caseε, it will be convenient to provide an agent in a εingle solution for adminiεtration.

If the agentε are amphoteric they may be utilized aε free baεes, as acid addition salts or as metal saltε. The εaltε muεt, of course, be pharmaceutically acceptable, and these will include metal salts, particularly alkali and alkaline earth metal salts, e.g., potasεium or sodium salts. A wide variety of pharmaceutically acceptable acid addition saltε are available. These include thoεe prepared from both organic and inorganic acidε, preferably mineral acidε. Typical acids which may be mentioned by way of example include citric, succinic, lactic, hydrochloric and hydrobromic acidε. Such products are readily prepared by procedures well known to those skilled in the art.

The agents (and inhibitorε) of the invention will normally be provided aε parenteral compoεitionε for injection or infuεion. They can, for example, be suspended in an inert oil, suitably a vegetable oil such as sesame, peanut, or olive oil. Alternatively, they can be suspended in an aqueous isotonic buffer εolution at a pH of about 5.6 to 7.4. Uεeful bufferε include sodium citrate-citric acid and sodium phosphate- phosphoric acid.

The desired isotonicity may be accompliεhed using sodium chloride or other pharmaceutically acceptable agents such as dextrose, boric acid, sodium tartrate, propylene glycol or other inorganic or organic solutes. Sodium chloride is preferred particularly for buffers containing sodium ions. If desired, solutions of the above compositions may be thickened with a thickening agent such as methyl cellulose. They may be prepared in emulsified form, either water in oil or oil in water. Any of a wide variety of pharmaceutically acceptable emulsifying agents may be employed including, for example acacia powder, or an alkali polyether alcohol sulfate or sulfonate such as a Triton.

The therapeutically useful compositions of the invention are prepared by mixing the ingredients following generally accepted procedures. For example, the selected components may be simply mixed in a blender or other standard device to produce a concentrated mixture which may then be adjusted to the final concentration and viscoεity by the addition of water or thickening agent and poεεibly a buffer to control pH or an additional solute to control tonicity.

For use by the phyεician, the compoεitionε will be provided in doεage unit form containing an amount of agent which will be effective in one or multiple doses to perform a desired function. Aε will be recognized by thoεe in the field, an effective amount of therapeutic agent will vary with many factorε including the age and weight of the patient, the patient'ε phyεical condition, the blood εugar level to be obtained, and other factorε. Adminiεtration Selected agentε, e.g., oligonucleotide or ribozymeε can be administered prophylactically, or to patients suffering from a target diseaεe, e.g. , by exogenouε delivery of the agent to an infected tiεεue by means of an appropriate delivery vehicle, e.g.. a liposome, a controlled releaεe vehicle, by use of iontophoresiε, electroporation or ion paired moleculeε, or covalently attached adducts, and other pharmacologically approved methodε of delivery. Routeε of adminiεtration include intramuεcular, aerosol, oral (tablet or pill form) , topical, εyεtemic, ocular, intraperitoneal and/or intrathecal. Expreεεion vectorε for immunization with ribozymeε and/or delivery of oligonucleotides are also suitable.

The εpecific delivery route of any εelected agent will depend on the use of the agent. Generally, a specific delivery program for each agent will focus on naked agent uptake with regard to intracellular localization, followed by demonstration of efficacy. Alternatively, delivery to these same cells in an organ or tiεεue of an animal can be purεued. Uptake εtudies will include uptake asεayε to evaluate, e.g., cellular oligonucleotide uptake, regardleεε of the delivery vehicle or εtrategy. Such aεεayε will also determine the intracellular localization of the agent following uptake, ultimately establiεhing the requirementε for maintenance of steady-state concentrations within the cellular compartment containing the target εequence (nucleuε and/or cytoplaεm) . Efficacy and cytotoxicity can then be teεted. Toxicity will not only include cell viability but alεo cell function.

Some methodε of delivery, e .cr. , for oligonucleotideε, that may be uεed include: a. encapsulation in liposomeε, b. transduction by retroviral vectors, c. conjugation with cholesterol, d. localization to nuclear compartment utilizing antigen binding εite found on moεt εnRNAs, e. neutralization of charge of oligonucleotides by using nucleotide derivatives, and f. use of blood stem cells to distribute oligonucleotideε throughout the body. At leaεt three typeε of delivery εtrategieε are useful in the present invention, including: agent modificationε, particle carrier drug delivery vehicles, and retroviral expression vectors. Unmodified agents may be taken up by cells, albeit slowly. To enhance cellular uptake, the agent may be modified esεentially at random, in ways which reduces its charge but maintainε εpecific functional groups. This results in a molecule which is able to diffuse acrosε the cell membrane, thus removing the permeability barrier.

Modification of agents to reduce charge is just one approach to enhance the cellular uptake of these larger moleculeε. The structural requirements necessary to maintain agent activity are well understood by those in the art. These requirements are taken into consideration when designing modifications to enhance cellular delivery. The modifications are also designed to reduce susceptibility to enzymatic degradation. Both of theεe characteristicε εhould greatly improve the efficacy of the agent.

Chemical modificationε of the phoεphate backbone of oligonucleotideε will reduce the negative charge allowing free diffuεion acroεε the membrane. This principle has been succesεfully demonεtrated for antiεenεe DNA technology. In the body, maintenance of an external concentration will be necessary to drive the diffusion of the modified oligonucleotides into the cellε of the tissue. Administration routes which allow the diseaεed tissue to be exposed to a transient high concentration of the oligonucleotide, which iε slowly disεipated by εyεtemic adsorption are preferred. Intravenous administration with a drug carrier designed to increase the circulation half- life of the oligonucleotides can be used. The size and compoεition of the drug carrier reεtrictε rapid clearance from the blood εtream. The carrier, made to accumulate at the εite of infection, can protect the oligonucleotides from degradative processes. Drug delivery vehicleε are effective for both systemic and topical administration. They can be designed to serve as a slow release reservoir, or to deliver their contentε directly to the target cell. An advantage of uεing direct delivery drug vehicleε is that multiple molecules are delivered per uptake. Such vehicles have been shown to increase the circulation half-life of drugs which would otherwise be rapidly cleared from the blood stream. Some examples of such specialized drug delivery vehicles which fall into thiε category are lipoεomeε, hydrogelε, cyclodextrinε, biodegradable nanocapεuleε, and bioadheεive microεphereε.

From this category of delivery syεtemε, lipoεomes are preferred. Liposomes increase intracellular stability, increase uptake efficiency and improve biological activity.

Lipoεomeε are hollow spherical vesicles composed of lipids arranged in a similar faεhion aε those lipids which make up the cell membrane. They have an internal aqueouε εpace for entrapping water εoluble compoundε and range in size from 0.05 to several microns in diameter. Several studieε have εhown that lipoεomeε can deliver agentε to cellε and that the agent remainε biologically active. For example, a lipoεome delivery vehicle originally deεigned aε a reεearch tool, Lipofectin, haε been εhown to deliver intact mRNA moleculeε to cellε yielding production of the corresponding protein.

Liposomes offer several advantages: They are non-toxic and biodegradable in composition; they display long circulation half-lives,* and recognition molecules can be readily attached to their surface for targeting to tiεεueε. Finally, coεt effective manufacture of lipoεome- baεed pharmaceuticalε, either in a liquid suspension or lyophilized product, has demonstrated the viability of this technology as an acceptable drug delivery εyεtem.

Other controlled releaεe drug delivery εyεtems, such aε nanoparticleε and hydrogelε may be potential delivery vehicleε for an agent. Theεe carriers have been developed for chemotherapeutic agents and protein-based pharmaceuticals. Topical administration of agents is advantageous since it allows localized concentration at the site of administration with minimal systemic adsorption. This simplifies the delivery strategy of the agent to the disease site and reduces the extent of toxicological characterization. Furthermore, the amount of material to be applied is far less than that required for other administration routes. Effective delivery requires the agent to diffuse into the infected cells. Chemical modification of the agent to neutralize negative or poεitive charges may be all that is required for penetration. However, in the event that charge neutralization is insufficient, the modified agent can be co-formulated with permeability enhancers, such as Azone or oleic acid, in a liposome. The liposomes can either represent a slow releaεe presentation vehicle in which the modified agent and permeability enhancer transfer from the lipoεome into the targeted cell, or the liposome phospholipids can participate directly with the modified agent and permeability enhancer in facilitating cellular delivery. In some caseε, both the agent and permeability enhancer can be formulated into a εuppoεitory formulation for slow release.

Agentε may alεo be εyεtemically adminiεtered. Syεtemic abεorption referε to the accumulation of drugs in the blood stream followed by distribution throughout the entire body. Administration routes which lead to systemic absorption include: intravenous, subcutaneous, intraperitoneal, intranasal, intrathecal and ophthalmic. Each of theεe adminiεtration routeε expoεe the agent to an acceεεible diεeased or other tisεue. Subcutaneous administration drains into a localized lymph node which proceeds through the lymphatic network into the circulation. The rate of entry into the circulation has been shown to be a function of molecular weight or size. The use of a liposome or other drug carrier localizes the agent at the lymph node. The agent can be modified to diffuse into the cell, or the liposome can directly participate in the delivery of either the unmodified or modified agent to the cell.

Moεt preferred delivery methodε include lipoεomeε (10-400 nm) , hydrogelε, controlled-releaεe polymers, microinjection or electroporation (for ex vivo treatments) and other pharmaceutically applicable vehicles. The dosage will depend upon the diεeaεe indication and the route of adminiεtration but εhould be between 10-2000 mg/kg of body weight/day. The duration of treatment will extend through the course of the diseaεe εymptomε, usually at least 14-16 days and posεibly continuouεly. Multiple daily doεeε are anticipated for topical applications, ocular applications and vaginal applicationε. The number of doεeε will depend upon diεease delivery vehicle and efficacy data from clinical trials.

Establiεhment of therapeutic levelε of agent within the target cell iε dependent upon the rate of uptake and degradation. Decreaεing the degree of degradation will prolong the intracellular half-life of the agent. Thus, chemically modified agents, e.g. , oligonucleotideε with modification of the phoεphate backbone, or capping of the 5' and 3' endε of the oligonucleotideε with nucleotide analogueε may require different dosaging.

It is evident from the above results, that by modulating telomerase activity and monitoring telomere length and telomerase activity, one may provide therapies for proliferative diεeaεeε and monitor the presence of neoplastic cellε and/or proliferative capacity of cells, where one is interested in regeneration of particular cell types. Asεayε are provided which allow for the determination of both telomere length, particularly aε an average of a cellular population, or telomerase activity of a cellular population. This information may then be used in diagnosing diseaseε, predicting outcomes, and providing for particular therapies.

All publications and patent applications cited in this specification are herein incorporated by reference as if each individual publication or patent application were specifically and individually indicated to be incorporated by reference.

Although the foregoing invention haε been deεcribed in some detail by way of illustration and example for purposes of clarity of understanding, it will be readily apparent to those of ordinary skill in the art in light of the teachings of this invention that certain changes and modifications may be made thereto without departing from the spirit or scope of the appended claims.

Claims

Claims 1. Method for treatment of a condition asεociated with an elevated level of telomerase activity within a cell comprising the step of: administering to said cell a therapeutically effective amount of an inhibitor of said telomerase activity.

2. Method for treatment of a condition asεociated with an increaεed rate of proliferation of a cell, comprising the step of: administering to said cell a therapeutically effective amount of an agent active to reduce losε of telomere length within said cell during said proliferation.

3. Method for extending the ability of a cell to replicate, comprising the εtep of: administering to said cell a replication extending amount of an agent active to reduce loεε of telomere length within εaid cell during cellular replication.

4. A pharmaceutical composition comprising a therapeutically effective amount of an inhibitor of telomerase activity in a pharmaceutically acceptable buffer.

5. A pharmaceutical compoεition compriεing a therapeutically effective amount of an agent active to reduce loεε of telomere length within a cell during proliferation of εaid cell, in a pharmaceutically acceptable buffer.

6. Method for diagnoεiε of a condition in a patient aεsociated with an elevated level of telomerase activity within a cell, comprising the εtep of: determining the presence or amount of telomerase within said cellε in εaid patient.

7. Method for diagnosis of a condition associated with an increased rate of proliferation in a cell in an individual, comprising the stepε of determining the length of telomereε within εaid cell.

8. Method for determining telomere length of an animal chromoεome or group of chromosomes, said method comprising: bringing together in a reaction mixture said chromosome(ε) or telomere compriεing fragment(ε) thereof, a primer having at leaεt two telomeric repeat unitε, and nucleoεide triphoεphateε having the εame nucleotideε aε the non-protruding εtrand of εaid telomere, wherein at leaεt one of εaid nucleoside triphosphates or primer is labeled with a detectable label; and a DNA polymerase; incubating said reaction mixture for sufficient time for said primer to be extended to provide a primer extended sequence; εeparating εaid primer extended by εize; and determining the size of said primer extended sequence by means of said label.

9. Method according to claim 8, wherein one of nucleoside triphosphateε iε labeled with a radioiεotope and εaid εize iε determined by the level of radioactivity in relation to the amount of DNA preεent.

10. Method according to claim 8, wherein said nucleosides are combinationε of A, T and C, or A, T and G.

11. Method of determining telomere length of an animal chromoεome or group of chromosomes, said method compriεing: fragmenting said chromosome (ε) by a reεtriction endonuclease having a four base recognition εite absent in the telomere sequence; bringing together said fragments and a primer for said telomeric sequence, wherein said primer iε labeled to allow for binding of said primer to a surface; crosε-linking εaid primer to εaid telomeric εequence; isolating said telomeric sequence by means of said label; and determining the size of εaid telomeric sequence bound to said surface.

12. Method according to claim 11, wherein said primer is conjugated with (1) an agent capable of crosε- linking nucleic acidε upon irradiation; and (2) a εpecific binding pair member; and said surface is conjugated with the complementary specific binding pair member.

13. Method according to claim 11, wherein said primer is conjugated with (1) an agent capable of crosε- linking nucleic acids upon irradiation; and (2) a particle.

14. Method of reducing the rate of telomere shortening in a proliferating cellular composition, said method comprising: introducing into cellε of εaid cellular composition primers having from 2 to 3 repeats of the repeating unit of the cellular telomere.

15. Method of measuring the telomerase activity of a composition, εaid method compriεing: combining 1 or more repeatε of the telomere unit sequence and nucleoside triphosphateε lacking cytidine nucleotide, wherein at least one of said primer or nucleoside triphosphates iε labeled with a detectable label, with the proviεo that when εaid compoεition lackε a telomere sequence complementary to said probe, εaid telomere εequence iε added to said composition; incubating εaid compoεition for a predetermined time for εaid primer to be extended to provide an extended εequence; and determining the rate of formation of said extended sequence.

16. Method according to claim 15, wherein one of said nucleoside triphosphateε is labeled with a radioisotope, and said determining is by measuring radioactivity per unit weight of DNA.

17. Method of inhibiting the proliferation of telomerase-compriεing immortalized cellε, εaid method comprising: contacting said immortalized cells with a telomerase inhibitor under conditions wherein said inhibitor enters said cells; whereby said proliferation of said cells iε inhibited.

18. Method according to claim 17, wherein εaid inhibitor inhibits expresεion of telomeraεe.

19. Method of according to claim 17, wherein said inhibition is an oligonucleotide sequence compriεing the complementary εequence of the telomeraεe RNA.

20. Method according to claim 17, wherein εaid inhibitor is a ribozyme.

21. Method for extending the proliferative capability of a mammalian cell population, said method compriεing: adding to said cells oligonucleotideε compriεing at least two repeats complementary to the sequence of the protruding strand of the telomere of the chromosomes of said cells, whereby the shortening of εaid telomere is slowed.

22. Method for treatment of a diseaεe or condition aεεociated with cell εeneεcenεe, comprising the stepε of: administering a therapeutically effective amount of an agent active to derepresε telomeraεe in the εeneεcing cellε.

23. Method for εcreening for a telomeraεe derepreεεion agent, compriεing the εtepε of: contacting a potential agent with a cell lacking telomeraεe activity, and determining whether εaid agent increaεeε the level of εaid activity.

24. The method of claim 23, wherein said cell is a cell expreεεing an inducible T antigen.

25. The method of claim 1, wherein εaid cell if a fungal cell, and εaid administering reduces viability of said cell.

26. The method of claim 25, wherein said cell is a C. albicans cell.

27. Method for εcreening for agentε uεeful in treatment of a human diεeaεe aεεociated with an elevated level of telomeraεe activity in a human cell, compriεing the step of testing potential said agents for activity in inhibiting telomerase activity.

28. Method for preparing cells from a donor for the reintroduction of said cellε to εaid donor, compriεing the εtepε of : isolating cellε from said patient, expanding said cellε in vitro, in the preεence of an agent which εlowε the replicative εeneεcence of εaid cellε, and introducing said cells after expanding into a patient.

29. Method of claim 28 wherein said cellε are εorted prior to expanding, and cellε with greateεt replicative capacity are iεolated.

30. Method of claim 29, wherein telomere length is used as a marker of replicative capacity.

31. Method of claim 28 wherein replicative seneεcence is εlowed by growing said cellε in a culture medium which εlowε replicative senescence.

32. Method of claim 31 wherein said culture medium contains a C-rich terminal sequence.

33. Method of claim 28 wherein said cells are CD4⁺ cellε.

34. Method of claim 33 wherein εaid method iε used for the treatment of AIDS .

35. Method of claim 28 wherein εaid cells are bone marrow stem cells .

36. Method of claim 28 wherein said cells are dermal fibroblasts .

37. Method of claim 36 wherein collagenase activity is used as a marker of replicative capacity.

38. Method of claim 36 wherein said method also compriεeε the εtepε of recombining εaid cellε with autologouε matrix proteinε obtained from εaid cellε .

39. Method of claim 2 wherein εaid agent iε telomeraεe.

40. Method of claim 39 wherein εaid telomeraεe iε iεolated from immortal human cellε .

41. Method of claim 40 wherein εaid immortal human cellε are human tumor cells.

42. Method of claim 41 wherein said cells are

U937 histiocytic lymphoma.

43. Method of claim 39 wherein εaid telomeraεe iε iεolated from Tetrahymena.

44. Method of claim 39 wherein εaid telomeraεe iε recombinant telomeraεe.

45. Method of claim 39 wherein εaid telomeraεe iε adminiεtered to the dermi .

46. Method of claim 45 wherein εaid telomeraεe iε adminiεtered by direct application.

47. Method of claim 39 wherein εaid telomeraεe iε adminiεtered to the endothelium.

48. Method of claim 39 wherein εaid telomeraεe iε adminiεtered to retinal pigmented epithelial cellε.

49. Method of claim 28 wherein εaid cellε are retinal pigmented epithelial cellε.

50. Method of claim 20 wherein said ribozyme is formed in a hammerhead motif.

51. Method of claim 20 wherein said ribozyme is formed in a hairpin motif.

52. Method of claim 20 wherein the RNA target for εaid ribozyme iε 3' AUCCCAAUC 5'.

53. Method of claim 28 wherein said cells are muscle sattelite cells.

54. Method of claim 53 wherein εaid method is effective for the treatment of muscular dystrophy.

55. Method of claim 1 wherein said cells are cancer cells.

56. Method of claim 55, wherein said inhibitor is an oligonucleotide consiεting eεεentially of 5'

TTAGGGTTAGGG 3' (SEQ ID NO: 3) .

57. Method of claim 6, wherein εaid cellε are cancer cellε, and εaid method detectε the preεence of εaid cancer cellε.

58. Method for the treatment of AIDS compriεing the following εteps: removing CD4⁺ cells from patient when AIDS is first detected, storing said CD4⁺ cells, and reintroducing cells to said patient at a time when patient'ε CD4⁺ cellε are limiting.

59. Method of claim 58 wherein said removed

CD4⁺ cells are expanded in vitro.

60. Method of claim 59 wherein said cells are expanded in the presence of agents which slow cellular seneεcence.

61. Method of claim 58 wherein telomere length of εaid patient' ε CD4⁺ cellε iε uεed aε a marker of the time when patient' ε CD4⁺ cellε are limiting.

62. Method of claim 2 or 3 wherein εaid agent iε administered by means of a liposome.

63. Method of claim 17, 20, or 22 wherein said telomerase inhibitor iε adminiεtered by meanε of a lipoεome.

64. Method of claim 6 wherein εaid telomeraεe activity iε determined by measuring the rate of elongation of an appropriate repetitive sequence having two or more repeats of the telomere unit sequence TTAGGG.

65. Method of claim 64 wherein εaid rate of elongation is determined by measuring the incorporation of labelled nucleoside triphoεphateε .

66. Method of claim 65 wherein εaid nucleoεide triphoεphateε are radioactively labelled.

67. Method of claim 65 wherein εaid nucleoεide triphoεphateε are labelled with fluoreεcein.

68. Method of claim 6 wherein εaid telomeraεe activity iε meaεured by the uεe of antibodieε εpecific for the telomeraεe protein.

69. Method of claim 6 wherein εaid telomerase activity is measured by the uεe of the polymeraεe chain reaction.

70. Method for identifying a bone marrow εtem cell compriεing the εtep of: detecting telomeraεe activity in a potential εaid cell, wherein the preεence of εaid activity is indicative of the presence of said cell.

71. Method for determining telomere length in a chromosome, comprising the step of: extending the end of a chromoεome with a defined oligonucleotide, annealing a primer to the junction formed between said defined oligonucleotide and the end of said chromosome, and extending said primer to produce an extension product; wherein the length of said extension product is a measure of said telomere length.