<%BANNER%>

Analysis of Rat Microglial Cellular Senescence as Determined by Measurements of Telomere Length and Telomerase Activity

University of Florida Institutional Repository

PAGE 1

ANALYSIS OF RAT MICROGLIAL CELLULAR SENESCENCE AS DETERMINED BY MEASUREMENTS OF TELOMERE LENGTH AND TELOMERASE ACTIVITY By BARRY ERIC FLANARY A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Barry Eric Flanary

PAGE 3

I dedicate this research to my beautiful wife, Allison.

PAGE 4

ACKNOWLEDGMENTS I thank my mentor, Dr. Wolfgang Streit, who welcomed me as a researcher into his laboratory, permitted me to pursue the area of research most compelling to me, and who taught me an enormous wealth of knowledge. I also thank Dr. Gerry Shaw, Dr. Satya Narayan, and Dr. Jeff Harrison for serving on my dissertation committee and for their scientific advice. Appreciation is also extended towards my fellow lab colleagues, including: Chris, Amanda, Josh, Parker, Tanya, Nicole, Au stin, Jackie, and Robert. I thank Dr. Michael Fossel, who has he lped open numerous doors of opportunity for me, for all of his help throughout the years, and for all that he has done for us, in particular for inviting me to be on the Editorial Board of his journal, The Journal of AntiAging Medicine, travelling to Illinois State Univers ity (when I was an M.S. graduate student there) to give a seminar on telomeres and cell senescence, and for inviting me to be a plenary speaker on his scientific panel at the Inaugural Inte rnational Convention on Longevity in Sydney, Australia in March of 2004. This thesis is written in honor of the la te President Ronald Reagan (who succumbed to Alzheimers disease while I was working on this dissertation research), with the hope that a cure for the disease can be found sooner rather than later. I especially thank my parents and brother for their support, and all that they have done to make it possible for me to pur sue a path of science in life. Special gratitude is expressed to my wife, Allison, for all of her patience, encouragement, and support while I was wo rking on this dissertation research. iv

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iv LIST OF FIGURES .........................................................................................................viii ABSTRACT ......................................................................................................................xii CHAPTER 1 INTRODUCTION........................................................................................................1 Structure and Function of Telomeres...........................................................................1 Characteristics of Cellular Senescence.........................................................................2 The Telomere Hypothesis of Cellular Aging................................................................3 Structure and Function of Telomerase..........................................................................4 Microglial Cells of the Central Nervous System..........................................................5 Microglial Structure and Function in vitro and in vivo .........................................5 Role of Microglia in Normal Brain Aging and Dementia.....................................7 Microglial Response Following Faci al Motor Nucleus Axotomy........................9 Neuroprotective Functions of Microglia.............................................................11 Specific Aims and Hypothesis....................................................................................12 2 PROGRESSIVE TELOMERE SHORTENIN G OCCURS IN CULTURED RAT MICROGLIA, BUT NOT ASTROCYTES................................................................14 Introduction.................................................................................................................14 Materials and Methods...............................................................................................14 Culturing of Microglia and Astrocytes................................................................14 Treatment of Microglial Cells.............................................................................16 Determination of Telomere Length.....................................................................16 Determination of Individual Chromosomal Telomere Length............................19 Determination of Telomerase Activity................................................................21 Determination of Cell Pro liferation and Viability...............................................23 Results.........................................................................................................................24 GM-CSF Stimulates Microglial Proliferation.....................................................24 Telomere Shortening Occurs in Cultured Rat Microglia....................................25 Three-Fold Variation Exists in In dividual Rat Microglia Telomeres.................30 Cyclical Telomere Shortening Occurs in Rat Astrocytes....................................32 Telomerase Activity in Cultured Rat Microglia and Astrocytes.........................38 Discussion...................................................................................................................41 v

PAGE 6

3 TELOMERES SHORTEN WITH AGE IN RAT CEREBELLUM AND CORTEX IN VIVO .....................................................................................................50 Introduction.................................................................................................................50 Materials and Methods...............................................................................................50 Collection of Rat Cerebellum and Cortex Tissues..............................................50 Determination of Telomere Length.....................................................................50 Determination of Telomerase Activity................................................................51 Results.........................................................................................................................51 Telomeres Shorten With Age in Rat Brain in vivo ..............................................51 Telomerase Activity in Ra t Cerebellum and Cortex...........................................51 Discussion...................................................................................................................57 4 AXOTOMY INCREASES TELOMERE LENGTH, TELOMERASE ACTIVITY AND PROTEIN IN AXOTOMY-ACTIVATED MICROGLIA...............................61 Introduction.................................................................................................................61 Materials and Methods...............................................................................................61 Rat Facial Nerve Axotomy..................................................................................61 FACS-Isolation of Rat Microglia from Micro-dissected Facial Nuclei..............62 Determination of Telomere Length.....................................................................63 Determination of Telomerase Activity................................................................63 Telomerase Western Blot Analysis.....................................................................63 Histochemistry.....................................................................................................64 Statistical Analysis of Data.................................................................................65 Results.........................................................................................................................65 Increase in Microglia Surroundi ng Axotomized Facial Nuclei...........................65 Increase in Telomere Length in Axotomized Facial Nuclei................................66 Increase in Telomerase Activity in Axotomized Facial Nuclei...........................69 Increase in Telomerase Protein Quantity in Axotomized Facial Nuclei.............72 FACS-Isolation of Microglia from Facial Nuclei...............................................74 Increase in Telomerase Activity in FACS-Isolated Microglia From Axotomized Facial Nuclei...............................................................................74 Discussion...................................................................................................................78 5 ALPHA-TOCOPHEROL (VITAMIN E) INDUCES RAPID, NON-SUSTAINED PROLIFERATION IN CULT URED RAT MICROGLIA.........................................85 Introduction.................................................................................................................85 Microglial Activation..........................................................................................85 Function of Vitamin E.........................................................................................86 Materials and Methods...............................................................................................87 Culturing of Microglia.........................................................................................87 Treatment of Microglial Cells.............................................................................87 Determination of Cell Proliferation.....................................................................88 Determination of Interleukin-1 Production.......................................................88 Determination of Telomere Length.....................................................................88 vi

PAGE 7

Determination of Telomerase Activity................................................................88 Statistical Analysis of Data.................................................................................88 Results.........................................................................................................................89 Microscopic Examination of Cultured Rat Microglia at Various Times and Treatments........................................................................................................89 Vitamin E Induces Cell Prolifera tion in Cultured Rat Microglia........................89 Telomere Length Analysis in Vitamin E-Treated Cultured Rat Microglia.........94 Telomerase Activity Analysis in Vita min E-Treated Cultured Rat Microglia....95 Interleukin-1 Beta Producti on in Cultured Rat Microglia...................................98 Discussion..........................................................................................................100 6 LIFE-SPAN EXTENSION IN NORMAL RAT MICROGLIA VIA TELOMERASE REVERSE TRAN SCRIPTASE RETROVIRAL TRANSDUCTION...................................................................................................107 Introduction...............................................................................................................107 Materials and Methods.............................................................................................107 Culturing of Microglia.......................................................................................107 Production of Replication-Defective Telomerase-Encoding Retroviruses.......107 Transduction of Rat Microglia With Telomerase-Encoding Retroviruses........109 Determination of Telomerase Activity..............................................................110 Telomerase Western Blot Analysis...................................................................110 Statistical Analysis of Data...............................................................................110 Results.......................................................................................................................110 Telomerase-Encoding Retroviral Vector...........................................................110 Telomerase Transduction Extends Life-Span of Microglia..............................111 Telomerase Activity in Transduced Microglia..................................................113 Telomerase Protein Quantity in Telomerase-Transduced Microglia................117 Discussion.................................................................................................................119 7 CONCLUSIONS AND IMPLICATIONS...............................................................125 Conclusions...............................................................................................................125 Implications..............................................................................................................128 LIST OF REFERENCES.................................................................................................130 BIOGRAPHICAL SKETCH...........................................................................................146 vii

PAGE 8

LIST OF FIGURES Figure page 2-1. Cell proliferation in GM-CSF-treated cultured rat microglia as determined by MTT analysis............................................................................................................25 2-2. Southern blot analysis for measurement of telomere length in cultured microglia...27 2-3. Telomere length dist ribution in control and GM -CSF-stimulated microglia on days 1, 16, and 32.....................................................................................................28 2-4. Southern blot analysis for measurement of telomere length in microglia cultured at varying densities...................................................................................................29 2-5. Telomere length distribution in micr oglia grown to near-confluence in various culture areas (9.5 cm2, 21 cm2, 175 cm2)...............................................................30 2-6. Telomere FISH analysis of metaphase spreads of cultured rat microglia using a FITC-conjugated peptide nucleic acid telomere-specific probe...............................31 2-7. Telomere fluorescence intensity (TFI) of all 168 individual telomeres in the 42 chromosomes of 2-day old cu ltures of rat microglia................................................32 2-8. Southern blot analysis for measur ement of telomere length in non-passaged astrocytes from day 1 to 10......................................................................................33 2-9. Telomere length distri bution in non-passaged astroc ytes from day 1 to 10..............34 2-10. Southern blot analysis for measurem ent of telomere length in non-passaged rat astrocytes from day 2 to 32......................................................................................35 2-11. Telomere length distribution in non-pass aged rat astrocytes from day 2 to 32.......36 2-12. Southern blot analysis for measuremen t of telomere length in astrocytes from passage 1 to 5...........................................................................................................37 2-13. Telomere length distribution in astrocytes from passages 1 to 5............................38 2-14. Telomerase activity in control (Con) and GM -CSF (CSF)-stimulated rat microglia on the indicated days................................................................................39 2-15. Telomerase activity in non-passaged rat astrocyt es on the indicated days..............40 viii

PAGE 9

2-16. Quantitation of telomerase activity (arbitrary units) in rat microglia and nonpassaged astrocytes on the indicated days................................................................41 3-1. Southern blot analysis for measurement of telomere restriction fragment (TRF) length in rat brain tissue...........................................................................................52 3-2. TRF length distribution in rat cerebellum and cort ex samples on days 21 and 152..53 3-3. Average TRF length in rat cerebellum and cortex tissues on days 21 and 152.........53 3-4. TRAP analysis for telomerase activ ity in rat brain tissue (days 21 to 182)..............54 3-5. Quantitation of telomerase activity (arb itrary units) in rat cerebellum and cortex tissues (days 21 to 182)............................................................................................55 3-6. TRAP analysis for telomerase activ ity in rat brain tissue (days 21 to 35)................55 3-7. Quantitation of telomerase activity (arb itrary units) in rat cerebellum and cortex tissues (days 21 to 35)..............................................................................................56 3-8. Overall, the cerebellum exhibits highe r telomerase activity than the cortex from day 21 to 35..............................................................................................................56 4-1. Micrographs of axotomized (A) and control (B) facial nucleus on day 3 postaxotomy stained with GSI-B4 le ctin to identify microglia......................................67 4-2. Southern blot analysis for measurement of telomere length in facial nuclei.............68 4-3. Densitometric quantitation of te lomere length in facial nuclei.................................69 4-4. Representative TRAP analysis imag e used for measurement of telomerase activity in facial nuclei.............................................................................................70 4-5. Densitometric quantitation of te lomerase activity in facial nuclei............................71 4-6. Densitometric quantitation of telomera se activity in unoperated facial nuclei.........72 4-7. Western blot image used for measurement of telomerase protein quantity in facial nuclei........................................................................................................................73 4-8. Densitometric quantitation of te lomerase protein in facial nuclei.............................73 4-9. FACS-isolation of microglia from axotomized and control facial nuclei.................76 4-10. TRAP analysis image used for meas urement of telomerase activity in FACSisolated facial nuclei.................................................................................................77 4-11. Densitometric quantitation of telome rase activity in FACS-isolated facial nuclei........................................................................................................................7 7 ix

PAGE 10

5-1. Representative micrographs of cultu red rat microglia under various treatment conditions.................................................................................................................90 5-2. Cell proliferation (as determined by M TT assay) of cultured rat microglia on the indicated days under various treatment conditions..................................................91 5-3. Proliferation rate (as determined by BrdU incorporation over 2 hours) of cultured rat microglia on the indicated days under various treatment conditions..................93 5-4. Proliferation of cultured rat mi croglia at 48 hours under various treatment conditions.................................................................................................................94 5-5. Densitometric quantitation of telomere length in cultured rat microglia on the indicated days under various treatment conditions..................................................95 5-6. Representative TRAP analysis imag e used for measurement of telomerase activity in cultured rat microglia..............................................................................97 5-7. Quantitation of telomerase activity in cultured microglia on the indicated days under various treatment conditions..........................................................................98 5-8. Quantitation of telomerase activity in cultured microgl ia at 48 hours under various treatment conditions....................................................................................99 5-9: Quantitation of interleukin-1 production by cultured rat microglia on the indicated days under various treatment conditions..................................................99 6-1. The Clontech retroviral vector, pLPC -hTRT, used to transduce cultured rat microglia with the human telomerase re verse transcriptase (i.e., hTRT) gene......112 6-2. Brightfield and green fluorescence micrographs of rat microglia and rat glioblastoma cells (RG-2) following retroviral transduction on day 4..................113 6-3. Representative micrographs of cu ltured rat microglia following retroviral transduction on days 1 and 20................................................................................114 6-4. Representative photographs of hTRT-t ransduced cultured rat microglia on days 57 and 75................................................................................................................115 6-5. TRAP analysis image used for measurem ent of telomerase activity of cultured rat microglia on the indicated days under various transduction conditions................116 6-6. Quantitation of telomerase activity in cultured rat microglia on the indicated days under various transduction conditions....................................................................117 6-7. Western blot image used for measurement of telomerase protein in cultured rat microglia on the indicated days under various transduction conditions................118 x

PAGE 11

6-8. Quantitation of telomerase protein quantity in cultured rat microglia on the indicated days under various transduction conditions............................................118 xi

PAGE 12

Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ANALYSIS OF RAT MICROGLIAL CELLULAR SENESCENCE AS DETERMINED BY MEASUREMENTS OF TELOMERE LENGTH AND TELOMERASE ACTIVITY By Barry E. Flanary May 2005 Chair: Wolfgang J. Streit Major Department: Neuroscience Normal somatic cells have a finite replic ative capacity, and with each cell division, telomeres (the physical ends of chromosomes) progressively shorte n until they reach a critical length, at which point the cells ente r cellular senescence. Some cells maintain their telomeres by the action of the telomera se enzyme. Microglia, a non-neuronal cell type residing within the centr al nervous system (CNS), play vital roles in maintaining neuronal function, health, and survival in both the normal and pa thological CNS. Microglia are the only adult cell type in the CNS that exhibit significant mitotic potential, suggesting that these cells have limited life-sp ans, may rely on proliferation to replace senescent cells, and are thus susceptible to telomere shortening and subsequent cellular senescence. In our studies, we have found that telome re shortening occurred in cultured rat microglia concomitant with their pr ogression to senescence by 32 days in vitro Telomere shortening also occurred in vivo in both rat cerebellum and cortex from day 21 xii

PAGE 13

to approximately 5 months of age (i.e., the oldest age analyzed). Axotomy-activated microglia from the facial nucleus (FN) maintained telomere length (TL) via increased levels of telomerase activity (TA) during periods of high proliferation in vivo Microglia isolated directly from the axotomized FN via fluorescence-activated cell sorting exhibited increased TA relative to un-operated controls suggesting that microglia are the primary cell type responsible for the increased TA obs erved in whole tissue FN samples. Vitamin E induced a significantly high prol iferation rate in cultured ra t microglia. This high rate of proliferation resulted in a concomitant de crease in TL, TA, and microglial activation. Microglia retrovirally-transduced with telo merase exhibited an increased maximal lifespan (ranging from 230 to 375%), and delayed en try into senescence, re lative to controls and empty-vector transduced microglia. Te lomerase transduction did not immortalize microglia, although these cells exhibited a normal phenotype, and had telomerase activity/protein present well past the time when all control cells had died. Our findings provide an impetus to furt her investigate rat microglial telomere dynamics in vivo especially with age, following a xotomy, or vitamin E supplementation, as well as in human microglia with age a nd incidence of Alzheimers disease. xiii

PAGE 14

CHAPTER 1 INTRODUCTION Structure and Function of Telomeres Telomeres are specialized structures at the physical ends of eukaryotic chromosomes consisting of e ssential proteins (e.g., TRF 1, TRF2, TIN2, tankyrase) and highly conserved repeated DNA sequences (Kipling, 1995; Shay, 1999). Telomeres control genes near chromosome ends (Wright and Shay, 1992) and may direct chromosome attachment to the nuclear me mbrane (Gottschling and Cech, 1984). The very ends of telomeres, which contain about 10 to 20 nucleotides of single-stranded DNA, form telomere loops (T-loops) by means of a single-strande d DNA invasion event, and are thought to protect chromosome ends from degradation and end-to-end fusions (Shay, 1999). Vertebrate telomeres compri se the same sequence of hexanucleotide repeats (TTAGGG) n (Moyzis et al., 1988). The length of telomeres is species-specific and ranges from 5 to 20 kilobases (kb) in humans (Harley et al., 1990) and from 20 to 150 kb in mice (Kipling and Cooke, 1990). Te lomere loss occurs with each round of DNA replication (Harley et al., 1990) due to the inability of DNA polymerases to completely replicate linear DNA molecules (Olovnikov, 1971, 1973, 1996; Watson, 1972), and may also occur as a result of oxida tive stress (von Zglinicki, 2002). Telomere length can be used as a predictor of the futu re replicative capacity of cells (Allsopp et al., 1992), and depends on both the age of the ce ll and the number of times the cell has already divided (Harle y et al., 1990). 1

PAGE 15

2 Characteristics of Cellular Senescence Normal somatic cells undergo only a finite number of cell divisions in vitro before entering a non-dividing state called cellular senescence (Hay flick, 1961). Senescence can also occur in replication-i ndependent manners, such as ac tivation of p53 pathways (Shay et al., 1991) or when sufficient cellular damage accumulates. Senescence, which ultimately culminates in cell death, is characterized by an irreversible arrest of cell proliferation (Hayflick, 1965), s ubstantial alterations in patter ns of gene expression (i.e., SAGE: Senescence-Associated Gene Expre ssion) (Bernd et al., 1982; Shelton et al., 1999; Funk et al., 2000), an incr easing resistance to apoptosis (Spaulding et al., 1999), cell-type specific changes in cell function and gene expression (Funk et al., 2000), and concomitant telomere shortening (Harley et al., 1990). The reducti on in proliferative capacity of cells from old donors (Bowman and Daniel, 1975) and patients with premature aging syndromes (e.g., Werner s yndrome, and Hutchinson-Gilford progeria syndrome) (Prokof'eva et al., 1982), as well as the accumulation of se nescent cells (Dimri et al., 1995) both in vitro and in vivo with altered patterns of gene expression (Shelton et al., 1999; Funk et al., 2000), implicates cellu lar senescence in aging and age-related pathologies (Fossel, 2000). Af ter cells have exhausted their replicative capacity, they reach their Hayflick Limit and become incapable of further division (Hayflick, 1965). In some cells, a progression of intracellular events can lead to crisis (Wright et al., 1989), which is characterized by the appearance of one or more critically-short telomeres (Allsopp and Harley, 1995), which activat e DNA-damaging signals (Harley, 1991), and cause end-to-end chromosomal fusions to occu r (Cui et al., 2002). Thus, at least in vertebrates, it seems that the s hortest telomere length, not th e average, is responsible for maintaining chromosome stability, cell viability, and determining when a cell will enter

PAGE 16

3 senescence (Hemann et al., 2001). At crisis, nearly all cells enter senescence and ultimately die by apoptosis (Payne et al., 1994), although some are able to up-regulate expression of the telomerase enzyme and become tumorigenic (de Lange, 1994). Following the senescence of a ce ll, replication of neighboring mitotic cells can occur, and their division to fill in the gaps left by se nesced cells may cause their own telomeres to shorten in the process. Thus, senescence can lead to a propagating cycle of accelerated aging among remaining cells (Fossel, 2000). The Telomere Hypothesis of Cellular Aging The telomere hypothesis of cellular aging proposes that telomere shortening in mitotic somatic cells contributes to and cause s their senescence, hastens the senescence of neighboring mitotic and post-mitotic ce lls (Harley et al., 1992), and underlies organismal aging (Fossel, 2000). This hypothe sis (Harley et al., 1992) suggests that if telomeres in somatic cells can be maintain ed at/above, or increased to, pre-senescent levels (e.g., via telomerase) in order to prev ent/reverse senescence, then replicative lifespan should increase as well (Wright et al., 1996a; Fossel, 1998). If the life-span of individual cells can be increased, then as a re sult, the life-span of the entire organism may also be increased (Harley et al., 1992). Thus if cell senescence can be slowed/prevented, then age-related diseases may also be slowed/prevented (Fossel, 1998, 2000). The presence of senescent cells may interfere with the normal functioning of, and may contribute to, organ and tissue aging (Dimri et al., 1995). Telomere shortening can be used as both an in vitro (Harley et al., 1990; Allsopp et al., 1992; Harley et al., 1992; Flanary and Streit, 2004) and in vivo (Lindsey et al., 1991; Kajs tura et al., 2000; Wright and Shay, 2002; Flanary and Streit, 2003) ma rker of cell replication and cell aging. Telomere shortening can cause changes in expr ession of genes nearest the telomere (i.e.,

PAGE 17

4 TPE: Telomere Position Effect) (Wright and Shay, 1992; Wood and Sinclair, 2002). Thus, as telomeres shorten with age, genes (especially those nearest the telomere) can get over-expressed. Telomere position effect can result in the age-related expression and/or over-expression of genes near a telomere that is dependent on both distance from the telomere and individual chromosomal telome re length. It provides a mechanism for the modification of gene expression that occurs throughout the re plicative life span of cells (Baur et al., 2001). The existence of TPE sugge sts that progressive loss of telomeres may lead to SAGE (Dimri et al., 1995; Fossel, 1998), which may affect both cell and organ function. Interestingly, some examples of human genes located nearest the telomere encode for well-known age-related diseases: cataracts, neuroblastoma, prostate cancer, Alzheimers disease, melanoma, obesity, colo rectal cancer, ovaria n cancer, diabetes, renal cell carcinoma, deafness, retinal de generation, Huntington disease, leukemia, coronary artery disease, breast cancer, os teoporosis, glaucoma, deafness, emphysema. Structure and Function of Telomerase Elongation of telomeres can occur by the action of the ribonucleoprotein enzyme telomerase, which adds tandem hexanucleotide (TTAGGG) n repeats de novo to 3 ends of mammalian telomeres using its own RNA as a template (Greider and Blackburn, 1985; Morin, 1989; Cech et al., 1997). Telomera se comprises two components: an RNA portion, which can be expressed in normal ce lls and is up-regulated during malignant transformation (Blasco et al., 1996), and a pr otein/catalytic portion, which is a reverse transcriptase expressed in r odent (Burger et al., 1997) a nd gametic/embryonic (Wright et al., 1996b) cell types, as well as during ma lignant transformation (de Lange, 1994). Telomerase can compensate for the conti nual shortening of telomeres that would otherwise occur in its absence. Elongation of telomeres can result in the extension of

PAGE 18

5 cellular life-span (Wright et al., 1996a; Bodnar et al., 1998) Many diverse types of normal human and animal cell types have been transduced and subsequently immortalized with telomerase, such as bovine adrenocortical cells (Thomas et al., 2000), endothelial cells (Yang et al ., 1999), epithelial ce lls (Bodnar et al., 1998), fibroblasts (Bodnar et al., 1998), keratinocyt es (Guo et al., 1998), lymphocytes (Hooijberg et al., 2000), myoblasts (Seigneurin-Ve nin et al., 2000), osteoblas ts (Yudoh et al., 2001), and pancreatic islet cells (Halvor sen et al., 1999). Reconstitution of telomerase (e.g., via retroviral transduction) in vitro into several diverse human and animal cell types can result in restoration of replicative potential, extension of telomere length and cellular life span, avoidance of cellular senescence (Bodnar et al., 1998; Vaziri and Benchimol, 1998), and reversion of gene expression to youthful levels (Funk et al., 2000) in the absence of tumorigenic changes (Belair et al., 1997; Jiang et al ., 1999; Morales et al., 1999; Harley, 2002). During central nervous system (CNS) development, telomerase is highly expressed in neural pr ogenitor cells, but sh arply decreases as synapses form, and when cells undergo apoptosis or differentiate (Kruk et al., 1996; Mattson and Klapper, 2001). Microglial Cells of the Central Nervous System Microglial Structure and Function in vitro and in vivo The central nervous system (CNS), whic h contains the brain and spinal cord, contain two main populations of cells: neurons and glia. Neurons are specialized cells important for relaying electrical signals to and from the brain and spinal cord. However, the majority of cells present within the CNS are not neurons, but glia. Glia (i.e., astrocytes, oligodendrocytes, microglia) provide structural, metabolic, and trophic support to neurons at all times. Microglia are distributed ubiquitously throughout the

PAGE 19

6 central nervous system (CNS), and func tion as resident macrophages and antigenpresenting cells of the CNS (Thomas, 1992). They have vital roles in supporting and maintaining neuronal function, health, homeosta sis, and survival in both the normal and pathological CNS microenvironment (Stre it, 2002a, b) by phagocytosing amyloid peptide (Frautschy et al., 1992) and secreting cytokines and neurotrophic factors (Streit et al., 1999; Nakajima et al., 2001; Streit, 2002a, b). Microglia have been aptly called the brains immune system because these cells share functional char acteristics of cells in the peripheral immune system (e.g., lymphocytes and macrophages) (Streit and Kincaid-Colton, 1995). In addition to originating from bone marrow-derived hematopoietic progenitor cells (Eglitis and Mezey, 1997; Hess et al., 2004), microglia are capable of expressing MHC antigens, Band T-cell lymphocyte markers, and other immune cell-specific antigens. Unlike astrocytes and oligodendrocytes, microglia are capable of significant division, especially following neuronal injury (Kreutzber g, 1966; Graeber et al., 1988, Svensson et al., 1994). Following acute CNS inju ry, there is rapid ac tivation of microglia and astrocytes. While acute microglial activ ation is marked by a conspicuous mitotic response, reactive astrocytes undergo primarily hypertrophy with markedly enhanced GFAP immunoreactivity, but show little mito sis (Graeber et al ., 1988; Graeber and Kreutzberg, 1986; Kreutzberg, 1996). Thus the mitotic ability of microglia in vivo is much greater than that of astrocytes. Inte restingly, when these g lial cell populations are maintained in vitro the mitotic potential of astrocyt es exceeds that of microglia, and astrocytes spontaneously form confluent monolayers that resemble those formed by cultured fibroblasts. Microglia, on the other hand, require stimulation with hematopoietic

PAGE 20

7 growth factors to undergo significant cell division in vitro (Giulian and Ingeman, 1988; Suzumura et al., 1990). However, th e mitotic potential of microglia both in vitro and in vivo suggests that these cells may rely on pr oliferation and self-renewal to replace senescent cells, and thus may ha ve limited cellular life-spans. Role of Microglia in Normal Brain Aging and Dementia The role of microglial cells in the aging CNS and in the development of age-related neurodegenerative disease remains unknown. Alzheimers disease (AD) is an agerelated, progressive neurological disorder characterized by significant memory loss, extracellular amyloid plaque deposition, intr acellular neurofibrillary tangle formation within neurons, loss of neuronal synapses the dysfunction and death of significant numbers of neurons, and memory loss (Mann an d Yates, 1981). AD currently afflicts 1 in 10 individuals over age 65 and nearly half of those over age 85, with the incidence rate doubling approximately every 4.4 years after age 60 (Kawas et al., 2000). Microglial cells are known to be clustered around amyloid beta (A )-containing senile plaques in the aged and AD brain (Itagaki et al., 1989), and this clustering likely occurs because the cells are gathering there in an attemp t to remove insoluble deposits of A (Frautschy et al., 1992). However, clearance of A is often not achieved, and this raises the possibility that the A clearing ability of microglia may be weakened or lost with aging. This may explain why substantial deposits of amyl oid plaques can be found in elderly nondemented individuals (Dickson et al., 1992). In addition, there ma y be overproduction of A such that microglia are overwhelmed by a larger-than-normal amyloid burden, which may compromise the ability of microglia to clear amyloid, and impair their other vital neuroprotective functio ns (Streit, 2002b).

PAGE 21

8 Studies conducted in post-mortem human brains have shown an increased incidence of microglial cytoplasmic structur al abnormalities (i.e., cytoplasmic swelling, twisted and shortened processes, and cyt oplasmic fragmentation) and dystrophy in the cerebral cortex of aged and AD-diseased br ains (Streit et al., 2004b), which support the hypothesis that microglia may become dysf unctional with age and that microglial dystrophy may contribute to their senescence, which in turn, may impair their neuronsustaining functions and ultimately lead to neur onal cell death. It is reasonable that the increased presence of dystrophic microglia in elderly individuals occurs because the cells ability to divide is declining as a re sult of aging (replicative senescence) thereby slowing the replacement of se nescent (dystrophic) microglia with younger cells. Based on these observations, we hypothesize that a redu ced ability of microglia to clear amyloid with age and incidence of AD may be the result of their cellular senescence. Understanding AD is complex and multi-faceted. Synapse loss is a hallmark characteristic of declining memory func tion with aging and may be linked to an impairment of neuronal and/or glial cell function Neuronal integrity and function, in turn, are highly dependent on the presence of fully functional glial cells. In the normal CNS, microglia are engaged in the continuous monitoring of neuronal well-being (Streit, 2002b). To ensure proper neuronal f unctioning, complex molecular and cellular interactions occur between ne urons and microglia. Since microglia are capable of producing both neuroprotective and neurotox ic molecules depending on the type of signals received from neurons (Streit et al ., 1999), any impairment in microglial function due to cellular senescence (or otherwise) co uld have profound consequences for neuronal activity and cognitive function in the normal aging brain. Over time, microglia may enter

PAGE 22

9 senescence and be less able, or unable, to ma intain neuronal health. As a result, when sufficient quantities of microglia have begun to senesce, the neurons they once supported may begin to degenerate, enter senescence, and ultimately die as well due to diminished glial support and maintenance. Neuronal ce ll death leads to loss of communication and synapses between neighboring neurons, and ultimately is the cause of memory loss evident with age and in AD. Thus, neur odegenerative changes may occur because microglia are becoming senescent and dysfunctio nal, and as a result, may inadvertently contribute to neurodegeneration due to impaired glial support. Neuronal cell death is a hallmark characte ristic of AD and may be linked to an impairment of microglial cel l function. Thus, understandi ng how microglia are involved in age-related deterioration of neuronal func tion is important for enabling the prevention of AD. A demonstration of microglial senescence with age would suggest that slow and progressive neurodegeneration and associated neuronal cell death, which are ultimately responsible for memory loss and dementia, may result from diminished or impaired microglial cell function. This could lead to the development of new drugs designed to enhance microglial cell function and/or to slow microglial telo mere shortening and senescence as potential treat ments of AD for humans. Microglial Response Following Facial Motor Nucleus Axotomy Even though the CNS is generally considered a post-mitotic tissue, it is important to note that microglia do retain a robust proliferative potent ial, especially under conditions of CNS injury (e.g., axotomy), as shown by DNA labelling studies using 3Hthymidine or bromodeoxyuridine (BrdU) (G raeber et al., 1988; Kreutzberg, 1966, 1996; Streit and Kreutzberg, 1988; Svensson et al ., 1994). During facial nerve axotomy, the facial nerve is cut outside th e brain and the reactions of f acial motor neurons and their

PAGE 23

10 glial environment can be studied in the br ainstem (Graeber et al., 1988; Kreutzberg, 1996). In the adult rat, unila teral axotomy of the facial nerve produces a robust, wellcharacterized microglial respons e within the ipsilateral faci al motor nucleus. Since the contralateral facial nucleus is surgically unaffected, it serves as an internal control. In adult rats, by approximately 4 weeks postaxotomy, the motor neurons of the facial nucleus regenerate their functional connectiv ity (Kreutzberg, 1996), as noted by regained whisker movement (Streit, 1996). In addi tion, axotomy of the facial nerve does not disturb the bloodbrain barrier (R aivich et al., 1998). After transection of the facial nerve, microglia but not astrocytes proliferate (Graeber et al., 1988), become hypertrophic, and express several cell surface molecules, such as complement receptor 3, major histocompatability complex (MHC) cl asses I and II (Streit et al., 1989), costimulatory molecules (e.g., B7-1) (De Simone et al., 1995), and several cell adhesion molecules (Moneta et al., 1993). Microglial cells are activated and increase in number in the facial nucleus following peripheral axotomy. Microglia become motile and migrate towards the injured motor neurons within th e axotomized facial nucleus, and microglial phagocytosis of bacteria can be observed in situ following axotomy (Schiefer et al., 1999). Within a few days following axotomy, mi croglia also maintain close contact with neurons and move along their dendrites, sugge sting a possible role for microglia in "synaptic stripping", the displa cement of afferent synaptic terminals from the motoneuron surface following axotomy (Kreutzberg, 1996; Schi efer et al., 1999). Under conditions of facial-nerve axotomy, facial motor neurons survive and will eventually regenerate their injured axons (Kreutzberg, 1996). Age does not affect the glial response to axotomy in regards to expression of glial fibrilla ry acidic protein (G FAP), leukocyte common

PAGE 24

11 antigen, type 3 complement receptor, and MHC classes I and II (Hurley and Coleman, 2003). The fact that microglia undergo prolifer ative bursts shortly af ter an acute injury suggests that mitosis affords a mechanism to provide greater numbers of microglial cells, and thus increased trophic support, during CN S injury and distress. However, their ability to divide also suggests that the lif e span of microglia may be limited, and makes them susceptible to replicative se nescence (Flanary and Streit, 2004). Neuroprotective Functions of Microglia Microglia play both neuropr otective and immunocompetent roles which serve to maintain neuronal health (Streit et. al., 1999; Streit, 2002b). Neurons are especially fragile cells, and their well-being and pr oper functioning are highly dependent on the presence of large numbers of microglia th at sustain a plethora of neuron-supporting functions. Microglia are ex tremely sensitive to even minor disturbances in CNS homeostasis and rapidly become activated and proliferate vigorously following nearly all neuropathologic conditions, such as nerve injury stroke, and trauma (Streit et al., 1999). Glial activation after injury is a beneficial and ostensibly necessary process, and serves not only to restore homeostasis within the CNS microenvironment but also to assist in the regeneration of injured neurons. In addition to protecting the CNS from invading microorganisms, microglia are important also for providing neuroprotection to normal and damaged neurons. Therefore, it is important to sustain a healthy microglia l population in order to help keep the CNS functioning properly. During times of increas ed stress (e.g., acute neuronal injury), microglia are especially important due to their unique ability to rapidly respond to neuronal injury via migration, proliferat ion and trophic factor production. The observation that there are as many microglia in the brain as there ar e neurons (Streit and

PAGE 25

12 Kincaid-Colton, 1995), in conjunction with the fa ct that microglia represent the only type of mature brain cells capabl e of undergoing mitosis and se lf-renewal, emphasizes the importance of these cells for providing cons tant monitoring of ne uronal well-being and targeted trophic support to neurons that may encounter acute stress situations. Following axotomy of motor a xons within the facial nerv e, neuronal survival and axonal regeneration is accompanied by vigorous microglial activation and cell proliferation. Thus, micr oglial activation, which begins long before axons have regenerated and serves to assist in the re generation of injured motor neurons, is an integral, and potentially crucial, compone nt of the regeneration process. Since axotomized motor neurons do regenerate, the rapid onset of microg lial activation likely occurs because injured neurons are recruiting nearby microglia to assist them in their struggle to survive and regenerate (Streit et al., 1999). These observations strongly support a neuroprotective and pro-regenerative role of microglia in the injured CNS. Specific Aims and Hypothesis The specific aims are as follows: 1. To determine if cultured rat microglia are subject to telomere shortening and senescence when cultured in vitro 2. To determine if telomere shortening occurs in the rat brain with aging. 3. To determine if neuronal injury-induced mi croglial proliferation within the facial nucleus resulted in telomere shortening in vivo 4. Investigate long-term effects of vitamin E in cultured rat microglia. 5. To determine whether exogenous delivery of the telomerase gene via retroviral transduction could prevent microglial senescence and extend lif e-span of cultured rat microglia. Collectively, these experiments have fo cused on studying the hypothesis that with aging, microglia undergo telomere shortening both in vitro and in vivo become

PAGE 26

13 increasingly dysfunctional, a nd ultimately enter cellular sene scence. The rationale for this hypothesis is based on the fact that microglia undergo cell division in vivo and are thus susceptible to telomere shortening with age. If this situation does indeed occur in vivo in multicellular te lomerase-negative organisms (e.g., humans), it may lead to a decline in microglial cell function with age, which in turn, would inhibit their ability to promote neuronal well-being. Thus, age-related neuron loss may be due to loss of microglial support.

PAGE 27

CHAPTER 2 PROGRESSIVE TELOMERE SHORTENIN G OCCURS IN CULTURED RAT MICROGLIA, BUT NOT ASTROCYTES Introduction To study the possibility that microglial cells in vitro are subject to replicative senescence, we decided to investigate telo mere shortening and telomerase activity in microglia. We now present evidence to show that progressive telomere erosion occurs in cultured rat microglia, while astrocytes exhib it a cyclical pattern of telomere shortening and lengthening. Materials and Methods Culturing of Microglia and Astrocytes Microglia were isolated from newborn Sprague-Dawley rat brains. The cerebral cortices of neonatal rats ( 3 days) were stripped of meninges and minced with a sterile scalpel blade in a 35 x 10 mm di sh containing filter-sterilized 37 C solution D (0.137 M NaCl, 0.2 M NaH 2 PO 4 0.2 M KH 2 PO 4 5.4 mM KCl, 5 mM dextrose (glucose), 58.5 mM sucrose, 0.25 g/mL Fungizone (Gibco, Carlsbad, CA), and 1 x 10 6 U penicillin/streptomycin in sterile water). The tissue fragments/cell suspension were incubated in 37 C solution D containing 1.0% trypsin (Invitrogen, Carlsbad, CA) for 30 min. at 37 C on a bi-directional tilting platform An equal volume of Dulbeccos modified Eagles medium (DMEM) containi ng 10% fetal bovine serum (Gibco, Carlsbad, CA) and 1% penicillin/streptomycin (complet e medium) was added to quench the trypsin reaction. The mixed brain cell susp ension was then passed through a 130 m Nitex filter 14

PAGE 28

15 (Tetko, Inc., Briarcliff Manor, NY) and centrifuged (4,000 rpm (2,900 g), 10 min). The resulting pellet was resuspended in 10 mL of complete medium, passed through a 40 m Nitex filter, and plated on poly-L-lysine (0.01 g/L) (Sigma-Aldrich, St. Louis, MO) coated, solution D-rinsed, 175 cm 2 flasks at a density of 1.5 br ains per flask. The cultures were incubated in complete medium at 37 C under 5% CO 2 After 4 days, the medium was changed and incubation was continued for an additional 3 days. Microglia were harvested from the whole brain cultures by shak ing the flasks on an orbital shaker (100 rpm) for 1 hour (which detached the looselyadherent microglia), a nd then collecting the medium containing the free-floating microglia. The cells were then pelleted from the medium by centrifuga tion (4,000 rpm (2,900 g), 10 min), resuspended in fresh complete medium, and immediately plated (day 0) in cell culture dishes at the appropriate cell concentrations as follows: 9.5 cm 2 plates (1.0 x 10 6 cells/well), 3.8 cm 2 plates (4.0 x 10 5 cells/well), or 0.32 cm 2 plates (3.4 x 10 4 cells/well). The optimal initial cell plating density was empirically determined in previous experiments. Cells were allowed to settle for 1 hour in a 5% CO 2 37 C incubator, and then the cu lture medium was changed to remove any contaminating non-adherent cells. The microglia were then treated with the appropriate concentration of a particular treatment regimen. Astrocytes normally form monolayers that cover the bottom of the culturing flask containing mixed brain cell cultures. To prep are enriched astrocyte cultures, all cells adhering to the astrocytic monolayer were de tached by vigorously shaking the flasks at 200 rpm for 1 hour. The culture medium containing the floating cells was then removed, and the remaining adherent astrocytes were rinsed with PBS, trypsinized, counted, and plated. Astrocytes were plated on day 0 at an initial density of 1x10 6 cells/well in 9.5

PAGE 29

16 cm 2 plates and allowed to divi de, or at a density of 2x10 6 cells in 175 cm 2 flasks and passaged when confluent. Treatment of Microglial Cells Microglia were treated on day 0 with either 1.0 nM (0.015 g/mL) or 10.2 nM (0.15 g/mL) recombinant rat granulocyte-macro phage colony stimulating factor (GMCSF) (R&D Systems, Minneapolis, MN), ), 100 nM lipopolysacc haride (LPS), or received no stimulation (control). In all ex periments, media (and respective treatment) were changed as needed (usually every 3 to 4 days). Determination of Telomere Length To measure telomere length (i.e., telomere restriction fragment (TRF) length: the length of the telomere plus sub-telomeri c DNA, the latter being dependent upon the particular cleavage sites of th e two restriction enzymes used ), Southern blot analysis using chemiluminescent detection and the DIG-High Prime DNA Labeling and Detection Starter Kit II (Roche Indianapolis, IN) was employ ed as previously described (Flanary and Streit, 2003, 2004), with minor modifications. Genomic DNA was isolated from 1) cultured cells at various time points, 2) whole brain tissue samples, or 3) pooled micro-dissected facial nuclei (approximately 7 mg wet weight each), using the DNeasy DNA isolation kit (Qiagen, Valencia, CA). DNA concentration was determined by absorbance at 260 nm, while DNA purity wa s calculated by the ratio of 260/280 nm absorbance, using a spectrophotometer. For Southern blotting, either 2.5 g (for cultured cells), or 5.0 g (for tissues) of DNA was dige sted with 10 units each of Hinf I and Rsa I overnight at 37 C. Following digestion, 5.0 L gel loading dye (0.25% xylene cyanol, 0.25% bromophenol blue, 30% glycerol) was ad ded to each sample. Each digested DNA

PAGE 30

17 sample, and two samples containing a digoxigenin-labeled molecular weight (MW) marker (100 ng each lane) (i.e., DIG DNA MW marker II) (Roche, Indianapolis, IN), were then loaded onto a horizontal 15 x 25 cm 0.5% agarose gel, and electrophoresed at 70 volts in 1X TBE buffer at 4 C with buffer recirculation until the bromophenol blue band ran off the gel and the xylene cyanol band reached 80% the length of the gel (approximately 21 hours). The gel was then soaked successively in depurination solution (0.25 M HCl) for 5 min, then denaturation solution (1.5 M NaCl, 0.5 M NaOH) for 2 x 10 min., and finally in neutralization soluti on (1.5 M NaCl, 0.5 M Tris) for 2 x 10 min. Unless otherwise noted, all incubations and wa shes were performed at room temperature with gentle agitation. The gel was rinsed thre e times in sterile doubledistilled water after each treatment noted above, and was then equilibrated in 20X SSC (3.0 M NaCl, 0.3 M sodium citrate; pH 7.0) for 10 minutes before the DNA was vacuum blotted (Boekel, Feasterville, PA) onto a positively-charged nylon membrane at 45 mbar for 45 min. Following UV-crosslinking (120 mJ/cm 2 ), a 2 min. wash in 2X SSC, and prehybridization at 41 C for 1 to 2 hours, hybridization of telomeric repeats was accomplished by using a digoxigenin-labeled telomere-specific oligonucleotide probe (TTAGGG) 3 Digoxigenin labeling of the probe (100 pmol) was accomplished by using the DIG oligonucleotide tailing kit (Roche, Indianapolis, IN). Following hybridization at 41 C overnight (approximately 17 hours) in a hy bridization oven (H ybaid, Franklin, MA) with gentle rotation, the membrane was washed in 2X wash buffer (2X SSC, 0.1% SDS) for 2 x 5 min., followed by washes in 0.5X wash buffer (0.5X SSC, 0.1% SDS) for 2 x 15 min. at 55 C in a hybridization oven with mode rate rotation. After washing the membrane in washing solution (0.1 M male ic acid, 0.15 M NaCl, 0.3 % (v/v) Tween-20;

PAGE 31

18 pH 7.5) for 2 min., the membrane was then so aked in blocking solution for 1 to 2 hours, followed by incubation with a 1:10,000 dilu tion of an alkaline phosphatase (AP)conjugated anti-digoxigenin antibody for 30 to 45 min. Following washes in washing solution for 2 x 15 min., chemiluminescent de tection was accomplished by washing the membrane in detection solution (0.1 M NaCl, 0.1 M Tris; pH 9.5) for 3 min., and then incubating the membrane (wrapped in plasti c wrap) in the AP-metabolizing substrate CDP-Star (Roche, Indiana polis, IN) for 5 min. at 37 C. Following exposure of the membrane to X-omat AR Film (Eastman Kodak Company, Rochester, NY) in an autoradiography cassette for 1 to 60 min., th e film was developed using a Konica SRX101A automatic film processo r (Konica Minolta, Mahwah, NJ). A digital image of the autoradiograph was generated by scanning it using a GS-710 calibrated imaging densitometer (BioRad, Hercules, CA). Telo mere lengths (shortest, mean, longest) of each sample were calculated by comparison to known MW standards present on the gel (in each outside lane), and quantified usi ng the computer progra m Telometric (version 1.2) (Grant et al., 2001). Telomere lengths (shortest, mean, longest) of each sample were determined as follows. The length of the shortest telome res represented the telomere signal (smear) on the autoradiograph correspond ing to the smallest telome res (lowest MW), which was calculated by measuring the bottom of the smear in each lane. Similarly, the length of the longest telomeres represented the signa l corresponding to the longest telomeres (highest MW), which was calculated by measuring the top of the smear in each lane. The length of the mean telomeres represented the signal corresponding to the average length of all telomeres with in the entire length of the smear. Telomere length

PAGE 32

19 measurement by current Southern blot techniques normally, and unavoidably, incorporates sub-telomeric DNA regions into the calculated MW due to the use of DNA digestion via restriction enzymes. Subtelomeric DNA regions are located directly adjacent to the telomere regions of DNA, and are still connected to the telomere region during gel electrophoresis due to the action of rest riction enzymes, which do not cleave directly at the telomere/sub-telomeric region. During gel el ectrophoresis, the presence of these small sub-telomeric DNA regions will sl ow the migration of telomere regions, which would otherwise migrate slightly faster in the absence of such attached subtelomeric regions. Thus, all protocols used fo r Southern blot for telomere length analysis utilizing restriction enzymes do not generate a true measur ement of actual telomere length, since the pure telomere DNA region (if unconnected to adjacent sub-telomeric regions) would migrate faster on the gel, and thus would have a slightly smaller actual MW than what was measured. Determination of Individual Chromosomal Telomere Length Fluorescence in situ hybridization (FISH) is a molecu lar cytogenic technique that is used to obtain information from metaphase (P oon et al., 1999) or interphase (De Pauw et al., 1998) cells, depending on the specific sequence of the fluorochrome-conjugated probe applied. We used a telomere-specifi c FITC-conjugated probe, and the binding of the probe to its target (telomeres) can be iden tified by a distinct green fluorescence signal at the tips of metaphase chromosomes. For FISH analysis, metaphase chromosomes were obtained from cultured microglia using standard methods. Briefly, fresh cultures (day 0) of microglia were stimulated to divide w ith 10 nM rrGM-CSF for 2 days on top of glass coverslips. Colchicine (10 g/mL) was added, and the cells were incubated for 1 hr. at

PAGE 33

20 37 C. This step disrupts and prevents forma tion of mitotic spindles, prevents completion of mitosis, and enriches the population of metaphase ce lls. Following aspiration, prewarmed (to 37 C) 0.075 M KCl was added, which make s nuclei swell osmotically and helps prevent chromosome overlap, and th e cells were incubated for 20 min. at 37 C. The cells were then fixed with ice-cold me thanol/acetic acid (3:1). A FITC-conjugated peptide nucleic acid (PNA) te lomere-specific probe (Dako, Carpinteria, CA) was added, which was used due to its high sensitivity and specificity. PNA is a synthetic DNA/RNA analog capable of binding 99 to 100% of telome re repeats. Additionally, this probe does not recognize subtelomeric sequences a nd, therefore, will allow for an exact measurement of telomere length. Chromo somes were counterst ained with 100 mg/mL propidium iodide and mounted with the antifa de reagent Vectashiel d. The preparations were viewed with both a Zeiss Axioskop 2 fl uorescence microscope connected to an RT color Spot digital camera (Diagnostic Instrume nts, Inc., Sterling Heights, MI) using a 60X and 100X oil lens, and a Bio-Rad 1040 ES confocal system connected to an Olympus IX70 inverted microscope using an Olympus planapo 60X 1.40 oil lens. High quality preparations were photographed usi ng a digital camera at 1024 x 1024 resolution. Using special software, telomere length was measured at the ends of individual chromosomes from digital images of metapha se spreads (Poon et al., 1999) using the Zeiss fluorescence microscope. Using ImagePro Plus software (Media Cybernetics, Carlsbad, CA), individual telomere fluoresce nce intensity was measured. This enabled a determination of whether intrachromosomal telomere length variation occurs in these cells, and if certain chromosomes are more susceptible to shortening over time.

PAGE 34

21 Determination of Telomerase Activity Telomerase activity was measured using the telomere repeat amplification protocol (TRAP) as previously described (Fla nary and Streit, 2003, 2004), with minor modifications. Total protein was isolated from 1) PBS-washed cultured cells, 2) whole brain tissue samples, or 3) pooled micro-di ssected facial nuclei (approximately 0.5 mg wet weight each), using 200 L CHAPS lysis buffer (5.0 mM -mercaptoethanol, 1.0 mM EGTA, 1.0 mM MgCl 2 0.1 mM phenylmethylsulfonyl fluoride, 10 mM Tris, 0.5% CHAPS, 10% glycerol). The protein extract solution was collected in RNAse-free tubes, incubated on ice for 30 min., and centrifuged at 12,000 g (14,000 rpm) for 20 min. at 4 C to sediment residual cell debris if present. Protein extrac ts were then aliquoted into RNAse-free tubes and stored at -80 C. Immediately prior to TRAP analysis, total protein concentration was measured using the BCA protein assay reagent (Pierce, Rockford, IL) as per the manufacturers recommended protoc ol. For TRAP analysis, each sample set included normal protein extracts, a telome rase-negative control (CHAPS lysis buffer, and/or RNAse-treated extracts: 10 mg/mL RNAse:sample (1:1) incubated for 20 min. at room temp.), and a telomerase-positive control (500 ng protein extract of a rat glioblastoma cell line RG-2). Each 50 L reaction initially contained 5.0 L 10X TRAP buffer (10 mM EGTA, 5 00 mM KCl, 15 mM MgCl 2 100 mM Tris, 1.0 mg/mL BSA, 0.05% Tween-20), 200 M dNTP (Roche, Indianapolis, IN), 100 ng telomerase substrate (TS) primer (5-AAT-CCG-TCG-AGC-AGA -GTT-3), 500 ng protein extract, and RNAse-free water up to 48 L. This mixture was incubated for 20 min. at room temperature to allow telomerase, if present and active, to add hexanucleotide telomeric repeats (i.e., TTAGGG) onto the 3 end of the TS primer, which is a substrate

PAGE 35

22 oligonucleotide and served as an artificial telomere. Following telomeric extension, 100 ng CX primer (5-CCC-TTA-CCC-TTA-C CC-TTA-CCC-TAA-3) and 5 units Taq polymerase (Fisher Scientific, Pittsburgh, PA ) were added. Telomere repeats were amplified by the polymerase chain reaction (PCR) using the TS (forward) and CX (reverse) primers, which generate a 40 base pa ir internal control band (i.e., TS-CX primer dimer) in each lane (including the negative control lane, since the presence of the internal control band is independent of the activity of telomerase ), and a ladder of products (generated by telomerase) containing 6-base increments beginning at 46 base pairs in telomerase-positive lanes. Both an increased quantity and intensity of bands present within the ladder of products co rrespond to an increased level of telomerase activity. If telomerase activity was absent, no product la dder was formed and only the internal control band was evident. PCR was carried out as follows: Initial denaturation at 94 C for 2 min. to inactivate telomerase, then 33 total cycles of the following: 94 C for 30 sec., 55 C for 30 sec., 72 C for 45 sec. Following PCR, 5 L filter-sterilized gel loading dye (0.25% xylene cyanol, 0.25% bromophenol blue, 50% glycerol, 50 mM EDTA) was added to each sample. The samples were then loaded onto a ve rtical 20 cm 12.5% nondenaturing polyacrylamide gel, and electrophores ed at 87 volts at room temperature in 0.5X TBE buffer until the bromophenol blue band ran off the gel and the xylene cyanol band reached 95% the length of the gel (approximately 21 hours). The telomerase products were visualized by staining the gel with a 0.01% solution of SYBR Green (Molecular Probes, Eugene, OR) for 40 min. in the dark with gentle agitation, and then photographing the gel under ultr aviolet light using an el ectronic gel documentation system (Gel Doc 2000, BioRad, Hercules, CA). Quantitation of telomerase activity (i.e.,

PAGE 36

23 the ladder of products formed in each lane) was performed using the densitometry computer program Quantity One (version 4.3.1) (BioRad, Hercules, CA). Average telomerase activity was determined by calculating the mean of individual quantitative measurements from two or more identical samples. Normalized telomerase activity was determined by comparing the average densitom etric values of identical samples run on different gels in order to make an accurate comparison of telomerase activity between all samples run on multiple gels. Determination of Cell Proliferation and Viability To assess cell proliferation, two methods were employed: the colorimetric MTT assay, and 5-Bromo-2-deoxyuridine (BrdU) incorporation. For the MTT assay (Boehringer Manheim, Indianapolis, IN), whic h measures both cell pr oliferation and cell viability, microglia were plated at an initial density of 3.4 x 10 4 cells/well in 0.32 cm 2 (96-well) plates. MTT labeling reagent (5 mg/mL of 3-[4,5-dimethylthiazol-2-yl]-2,5diphenyl tetrazolium bromide in PBS) (Boe hringer Manheim, Indianapolis, IN) was added (10 L per well) to the cultured cells at various time points. Metabolically-active cells cleave the yellow tetrazolium MTT sa lt to form purple formazan crystals via NADH reductase. Following a 4 hour incubation at 37 C under 5% CO 2 cells were solubilized overnight in 10% SDS in 0.01 M HCl (100 L per well). The solubilized formazan product was spectrophotometrically quantified at 550 nm (using a reference wavelength of 655 nm) using a Benchmark microplate read er (BioRad, Hercules, CA) and microplate manager software (version 4.0). To determine proliferation using BrdU incorporation, microglia were plated at an initial density of 2.0 x 10 5 cells/well in 1.9 cm 2 plates (24-well plates). BrdU (Sigma

PAGE 37

24 Aldrich, St. Louis, MO; cat alog # B5002) was added (10 M final concentration) to the cultured cells at various time points. Prolif erating cells incorporated the BrdU (in place of thymine) during S-phase. Cells were fixed at 4 C overnight in 80% EtOH. The following morning, wells were rins ed with PBS and stored at 4 C in PBS until assayed. For BrdU analysis, each plate wa s incubated for 10 minutes at 37 C in 2M HCl. Following washes in PBS for 3 x 5 min., bl ocking buffer (PBS containing 0.1% Triton X100 and 2% normal goat serum) was added, and each plate was incubated for 30 min. at 37 C. A FITC-conjugated rat anti-BrdU an tibody (Serotec, Raleigh, NC; catalog # MCA-2060-FT) diluted in blocking buffer was added to each well, and each plate was incubated for 2 hours at room temperature c overed with foil (to prevent degradation of fluorescent signal). Wells were rinsed with PBS, and cell nuclei were stained with a 1.0 g/mL solution of 4'-6-Diamidino-2-phenylindol e (DAPI) for 5 min. covered with foil. Cells were photographed under fluorescence using a digital camera (Sony DSC-S75 Cyber-shot, 3.3 megapixels, Carl Zeiss Vario-Sonnar lens) conn ected to a Zeiss Axiovert 25 fluorescence inverted microscope. Cell viability was performed on cultured microglia using the live/dead viability/cytotoxicity kit (Mol ecular Probes, Eugene, OR). Results GM-CSF Stimulates Microglial Proliferation Microglia exposed to GM-CSF increased pro liferation compared to controls (Fig. 2-1). Cell viability analysis of cultured micr oglia indicated that with increasing time in culture, the total number of viable cells d ecreased in both control and GM-CSF-treated groups, indicating that the cells were enteri ng replicative senescence (data not shown).

PAGE 38

25 Figure 2-1. Cell proliferation in GM-CSF-treated cultured rat microglia as determined by MTT analysis. Following stimulation with the mitogen rrGM-CSF (CSF), microglia undergo a signifi cant burst (p<0.001) in pro liferation from day 2 to 9. However, when cultured in the co ntinual presence of CSF, microglia undergo rapid telomere shortening (Fla nary and Streit, 2004), and exhibit a significant decrease (p<0.001) in proliferation (when comparing day 9 to 41, day 16 to 41, or day 27 to 41). Importantly, proliferation in CSF-stimulated microglia eventually falls to levels belo w that evident in un-stimulated control microglia (i.e., on day 41). This suggests that microglia in vitro undergoing continual rapid division apparently use up most of their replicative potential within the first 14 days (as evidenced by the smaller burst in proliferation from day16 to 27 (p>0.05) compared to day 2 to 9), and are no longer able to sustain this rate at later timepoints, at which point they likely succumb to replicative senescence. Telomere Shortening Occurs in Cultured Rat Microglia Both control and rrGM-CSF (GM-CSF)-stim ulated microglia underwent telomere shortening while in culture (Fig. 2-2). In control microglia, both the longest and mean telomere length remained relatively unchanged from day 1 to 16 while the shortest telomeres decreased a moderate 6.3 kb. In contrast, GM-CSF-stimulated microglia exhibited dramatic telomere loss (10.8 kb) in th e longest telomeres from day 1 to 16, with

PAGE 39

26 minimal shortening (0.9 kb) occurring from day 16 to 32. Mean telomere length decreased only slightly over time. Convers ely, the quantity of s hort telomeres greatly increased (9.1 kb) from day 1 to 16, with a small increase (2.1 kb) occurring from day 16 to 32 (Fig. 2-3). Thus, in stimulated micr oglia, the longest telomeres were allowed to shorten while short telomeres were lengthene d. Additionally, the s hortest telomeres are continually being lengthened over time, even immediately pr ior to senescence (around day 32). The quantity of longest telomeres was much larger (9 kb) in GM-CSF-treated microglia than in controls on day 1 (Fig. 2-3). This indicates that rapid telomere lengthening occurs in microglia during the fi rst 24 hours of being exposed to GM-CSF, and is supported by TRAP data which shows that telomerase activ ity is substantially higher in stimulated microglia, compared to unstimulated microglia. No DNA was available to be isolated on day 32 in control cells, and very little in GM-CSF-treated cells, since they were already senescent at this time point. When microglia were plated in various sized culture dishes and grown to confluence, telomere length diminished proportional to the size of the culture dish. With increasing culturing area (i.e., 9.5 cm2, 21 cm2, and 175 cm2), telomere length decreased in the longest (1.4 kb), mean (1.3 kb), and shortest (2.1 kb) te lomeres (Fig. 2-4, 2-5).

PAGE 40

27 Figure 2-2. Southern blot analysis for m easurement of telomere length in cultured microglia. Genomic DNA was isolated on the indicated days (1, 16, 32) in cells treated with either sterile wate r (CON), or 10 nM rrGM-CSF (CSF). For both treatments, telomere length decreas ed over time. However, telomere shortening was much more pronounced in GM-CSF-treated cells.

PAGE 41

28 0 5 10 15 20 25 30 35 40 45Telomere Length (kb) Longest 32.933.441.931.130.2 Mean 29.629.432.730.229.0 Shortest 22.316.014.423.526.6 CON 1CON 16CSF 1CSF 16CSF 32 Figure 2-3. Telomere length distribution in control and GM-CSF-stimulated microglia on days 1, 16, and 32.

PAGE 42

29 Figure 2-4. Southern blot analysis for m easurement of telomere length in microglia cultured at varying densities. Cells we re plated without GM-CSF at varying densities and culture areas as follo ws: 5x105 cells (9.5 cm2), 5x105 cells (21 cm2), or 2x106 cells (175 cm2). Genomic DNA was isolated once cells reached near-confluence following 3 days of growth. Increased telomere attrition is evident in microglia plated in larger culture areas evidently caused by their increased division, which was neces sary in order to fill the available culturing area.

PAGE 43

30 0 5 10 15 20 25 30 35Telomere Length (kb) Longest 31.0 30.6 29.6 Mean 29.3 29.1 28.0 Shortest 25.8 26.9 23.7 9.5 21 175 Figure 2-5. Telomere length distribution in microglia grown to near-confluence in various culture areas (9.5 cm2, 21 cm2, 175 cm2). Three-Fold Variation Exists in I ndividual Rat Microglia Telomeres Individual microglia telomere s were visualized using FI SH analysis (Fig. 2-6). Telomere fluorescence intensity (TFI) is a meas urement (arbitrary unit s) used to provide an indicator of the number of telomeric re peats present on a particular telomere. Microglia on day 2 of in vitro growth were found to have a normal karyotype consisting of 42 chromosomes. The TFI was found to vary greatly both between and within individual chromosomes (Fig. 2-7). Variation in TFI values on all chromosomes ranged from 17.0 to 52.0, representing a 3.1-fold maximal difference in telomere length (mean TFI = 44.2 5.2). Variation of TFI values on th e same chromosome ranged from 17.6 to 47.0, representing a 2.7-fold difference in telomere length.

PAGE 44

31 Figure 2-6. Telomere FISH analysis of meta phase spreads of culture d rat microglia using a FITC-conjugated peptide nucleic acid telomere-specific probe. Cells were viewed at 90X (left) and 270X (right) magnification via confocal microscopy. Note the green telomeres at the ends of red chromosomes. Two large red cell nuclei containing interphase chromosome s are visible in the upper portion of the 90X view. Substantial telomere le ngth variation was found to exist both between and within individual micr oglial telomeres. Scale bar is approximately 10 m (in 90X view) and 5 m (in 270X view).

PAGE 45

32 Figure 2-7. Telomere fluorescence intensity (T FI) of all 168 individual telomeres in the 42 chromosomes of 2-day old cultures of rat microglia. TFI is an arbitrary measurement corresponding to the numb er of telomeric repeats present on each telomere. Thus, TFI is directly proportional to telomere length. Note the large quantity of long telomeres, as evidenced by high TFI values, which are present in young microglial cultures. Cyclical Telomere Shortening Occurs in Rat Astrocytes Astrocytes were also found to undergo telomere shortening during the first 10 days of in vitro culturing (Fig. 2-8). From day 1 to 3, telomere length decreased rapidly, with the differences being 12.7 kb in the l ongest, 6.1 kb in the mean, and 13.8 kb in the shortest telomeres. Beginning on day 3, a fluc tuating pattern in telomere length in all groups (i.e., longest, mean, shorte st) begins to develop. From day 3 to 4, telomeres in all

PAGE 46

33 groups (except the longest) are lengthened, most notably the shortest (10 kb). However, from day 4 to 5, telomeres in all groups shorte n. From day 5 to 7, telomeres in all groups are again lengthened. Then, from day 7 to 10, telomeres in all groups (except the longest from day 8 to 9) shorten again (Fig. 2-9). Figure 2-8. Southern blot analysis for meas urement of telomere length in non-passaged astrocytes from day 1 to 10. Telomere length decreased over time, especially during days 1 to 3. A cyclical patte rn of telomere shortening and relengthening was observed from day 1 to 10.

PAGE 47

34 0 5 10 15 20 25 30 35 40 45 50Telomere Length (kb) Longest 49.641.036.934.233.634.636.034.435.033.9 Mean 38.833.430.731.931.431.833.633.031.130.4 Shortest 32.926.219.129.126.629.430.630.428.226.9 12345678910 Figure 2-9. Telomere length distribution in non-passaged astrocytes from day 1 to 10. Note the cyclical pattern of telomere lengthening and shortening that occurs with time in culture. During lengthening, the shortest telomeres are lengthened th e quickest, suggesting that maintenance of the shortest telomeres is most critical (Hemann et al., 2001). Astrocytes were also found to have consistently l onger telomeres in all groups compared to microglia. Mean telomere length in astrocyt es on day 1 is 6.1 kb longer than in GMCSF-treated microglia, and 9.2 kb longer than in control microglia on the same day. The longest telomeres in astrocytes on day 1 are 7.7 kb longer than in GM-CSF-treated microglia, and 16.7 kb longer than in control microglia on the same day. Similarly, the shortest telomeres in astrocytes on day 1 ar e 10.6 kb longer than in control microglia, and 18.5 kb longer than in GM-CSF-treated cells on the same day (Fig. 2-3 and 2-9). Long

PAGE 48

35 term growth of cultured astroc ytes (up to day 32) revealed a similar pattern of telomere lengthening followed by a period of telomere erosion (Fig. 2-10). From day 2 to 16, telomere length increased in the longest ( 4.9 kb), mean (4.1 kb), and shortest (1.1 kb) telomeres. Similarly, from day 16 to 32, telo meres shorten in the longest (3.5 kb), mean (3.0 kb), and shortest (1.2 kb) telomeres (Fig. 2-11). This cyclical pattern of telomere lengthening and shortening in as trocytes is suppor tive of a similar cyclical pattern of transient telomerase expression and repressi on that is occurring in these cells. Figure 2-10. Southern blot analysis for m easurement of telomere length in non-passaged rat astrocytes from day 2 to 32. Telo mere length increased from day 2 to 16, then declined from day 16 to 32. This cyclical pattern of telomere lengthening and attrition correlates to increasing a nd decreasing telomerase activity over the same time period.

PAGE 49

36 0 5 10 15 20 25 30 35Telomere Length (kb) Longest 27.229.532.128.228.6 Mean 25.727.529.826.926.8 Shortest 24.325.325.417.724.2 29162432 Figure 2-11. Telomere length distribution in non-passaged rat astrocytes from day 2 to 32. Note the cyclical pattern of telo mere shortening and lengthening that occurs with time in culture. When astrocytes are passaged serially mean telomere length remains nearly unchanged from passage 1 to 5 (Fig. 2-12). However, both the longest and shortest telomeres undergo a cyclical pattern of telome re lengthening and shortening. Both sets of telomeres lengthen from passage 1 to 2, and then shorten from passage 2 to 3. Subsequently, they re-lengthen from passage 3 to 4, and then shorten from passage 4 to 5. The shortest telomeres exhibit the most drama tic lengthening. From passage 1 to 2 they lengthen 7.4 kb, and from passage 3 to 4 they lengthen 9.7 kb. The overall trend for the shortest telomeres is a continual lengthening over time (Fig. 2-13).

PAGE 50

37 Figure 2-12. Southern blot analysis for m easurement of telomere length in astrocytes from passage 1 to 5. The interval be tween each passage was as follows: passage 1 to 2 (4 days), 2 to 3 (6 days), 3 to 4 (6 days), 4 to 5 (9 days). While the mean telomeres remain nearly unchanged, the longest and shortest telomeres undergo a cyclical pattern of telomere lengthening and shortening, with the shortest telomeres gradually lengthening over time.

PAGE 51

38 0 5 10 15 20 25 30 35Telomere Length (kb) Longest 29.431.629.831.028.3 Mean 27.027.727.727.727.2 Shortest 11.819.217.627.326.7 12345 Figure 2-13. Telomere length distribution in astrocytes from passages 1 to 5. Note the cyclical pattern of telomere shorteni ng and lengthening that occurs with time in culture. Telomerase Activity in Cultured Rat Microglia and Astrocytes Telomerase activity was measured in bot h microglia and astrocytes on various days (Fig. 2-14 to 2-16). Telomerase activity was consistent, and at a low level, in control microglia from day 0 to 32. In contrast, GM-CSF-stimulate d microglia express a nearly 3-fold increase in telomerase activity on day 2, compared to controls. Subsequently, telomerase activity follows a cyclical pattern and declines until day 24, then increases dramatically again from day 24 to 32 (Fig. 2-14 and 2-16). In astrocytes, telomerase activity gradually increased from day 0 to 2, then rapidl y increased from day 2 to 16. Telomerase activity then dramatica lly declined from day 16 to 24. By day 32, activity was nearly non -existent (Fig. 2-15, 2-16). Telo merase activity correlated well

PAGE 52

39 with telomere length in astrocytes. Mean telomere length was found to increase in astrocytes by 4.1 kb from day 2 to 16, corre sponding to a period of high telomerase activity, especially on days 9 to 16 (during which time telomeres were lengthened). Subsequently, mean telomere length decrea sed 3.0 kb from day 16 to 32, corresponding with a period of low/absent telomerase activ ity (during which telomeres shortened) (Fig. 2-11 and 2-16). Figure 2-14. Telomerase activity in cont rol (Con) and GM-CSF (CSF)-stimulated rat microglia on the indicated days. Telo merase activity in control microglia remains relatively consistent. However, in GM-CSF-stimulated microglia, telomerase activity is increased drama tically on days 2 and 32, compared to controls. (-) = telomerase-negative control; (+) = telomerase-positive control; IC = internal control.

PAGE 53

40 Figure 2-15. Telomerase activity in non-passa ged rat astrocytes on the indicated days. Telomerase activity in astrocytes exhibits a cyclical pattern of increasing and decreasing telomerase activity. (-) = telomerase-negative control; (+) = telomerase-positive control; IC = internal control.

PAGE 54

41 0 10 20 30 40 50 60 70 80 90 01020304 DayTelomerase Activity 0 Astrocyte Microglia Con Microglia CSF Positive Con Figure 2-16. Quantitation of telomerase activity (arbitrary units) in rat microglia and non-passaged astrocytes on the indicated days. Control microglia express low telomerase activity than the positive control. GM-CSF-stimulated microglia exhibit a cyclical cycle of increasi ng and decreasing telomerase activity. Telomerase activity in astrocytes on day 0 is nearly 3-fold higher than in microglia on the same day and was alwa ys higher than the positive control. Additionally, GM-CSF-stimulated microglia exhibit a 3-fold higher level of telomerase activity on day 2, compared to controls. Discussion We have shown that both microglia and astrocytes undergo dynamic changes in both telomere length and telomerase activity with time in culture. While telomere shortening occurs gradually in actively di viding microglia and is accompanied by their progression into senescence, astrocytes exhib it a cyclical pattern of telomere lengthening and shortening and are able to divi de for much longer periods of time in vitro than microglia. Our study, which is the first to report on telomere length and telomerase

PAGE 55

42 activity in glial cells of any organism, show s that telomeres shorten progressively in microglia with increased cell division and with time in culture. The fact that microglia can be induced to proliferate following CNS injury suggests that microglia represent a se lf-renewing population of cells. Tour current findings open the possibility that the replicative potential of microglia in vivo is limited and that these cells may at some point exhaust their replicat ive capacity. Thus, injury-induced mitosis of microglia in vivo could result in telomere shorte ning, which may drive the cells down an accelerated path towards cellular senescence. Rapid elongation of the shortest telomere s in GM-CSF-treated cells occurs from day 1 to 16, suggesting that these cells are trying to evade senescence by maintaining the shortest telomeres above the critical length. The shortest telomeres lengths in GM-CSFtreated cells are 14.4 kb on day 1, while contro ls have lengths of 16 kb on day 16 (and enter senescence shortly thereafter). Thus, the critical telomere length that triggers senescence appears to be somewhere sligh tly below 14.4 kb. Control microglia exhibit telomere attrition of th e shortest telomeres, while the l ongest and mean telomeres remain nearly unchanged, from day 1 to 16, which sugge sts that either the longest telomeres are being preferentially maintained, or that non-uniform telomere lengthening occurs such that the longest telomeres actua lly increase in size (by 0.5 kb) while the shortest continue to erode (by 6.3 kb). The data al so suggests that attrition of the shortest telomeres occurs in the presence of low levels of telomerase (since both longest and mean telomeres are maintained). This is supported by our TRAP da ta indicating that telomerase is present in low amounts, which is likely enough to mainta in the longest and mean telomeres, while the shortest telomeres continually erode. Un stimulated cells appear to enter senescence

PAGE 56

43 when telomeres reach critical lengths, howev er, the GM-CSF-stimulated microglia do not. No critical length is reached in thes e stimulated cells, yet they still undergo senescence, as evidenced by reduced mitotic act ivity and viability after several weeks in culture, suggesting that mechanisms other than critically-short telomeres (e.g., ROS damage, cellular trauma, shortening of th e longest telomeres) may be triggering senescence in these cells in th e absence of telomerase repres sion (Kodama et al., 2001). Telomere erosion is also thought to occur directly by other means, which are independent of cell replication (von Zglinic ki, 2002). Since GM-CSF-treated microglia on day 1 have longest telomeres that are 9 kb longer than in controls on the same day, GM-CSF may up-regulate high te lomerase activity during the first few days of growth (Szyper-Kravitz et al., 2003), which is likely to prepare th e cells for the rapid division that subsequently occurs. This idea is s upported by our TRAP data, which shows greatly increased telomerase activity in microglia from day 0 to 2 (i.e., nearly 3-fold higher, compared to controls). Following day 1, GM-CSF treatment induces only maintenance of the shortest telomeres, as indicated by their continua l lengthening until senescence, while the longest and mean telomeres are a llowed to shorten. After day 2, telomerase activity in GM-CSF-stimulated microglia stead ily declines until day 24, then increases dramatically again to day 32. This supports our Southern blot data, indicating that telomeres continue to be lengthened until se nescence. The elongation of short telomeres (by 9.1 kb) in GM-CSF-stimulated microglia fr om day 1 to 16 corresponds with a nearly 3-fold higher increase in telomerase activity during this time. Similarly, short telomeres increase by 3.1 kb from day 16 to 32, corresponding to an approximate 2-fold increase in telomerase activity during this time. It is apparent that the rate of elongation of short

PAGE 57

44 telomeres (relative to measured telomerase activity) from day 16 to 32 is lower compared to day 1 to 16. At day 32, nearly all microg lia are likely senescent, and thus may not be able to lengthen telomeres as efficiently or recruit telomerase as proficiently as on day 2, suggesting that microglia are less able to util ize available telomerase with increasing time in culture. The level of telomerase activ ity is likely enough to maintain the shortest telomeres while the longest and mean telo meres continually erode. Undergoing rapid division, as during GM-CSF treatme nt, appears to enable microglia to rapidly increase telomere length initially, then on subsequent days to recruit telomerase to the shortest telomeres, while allowing the longest and mean telomeres to shorten. The opposite is seen in control microglia, in which only the shortest telomeres erode. Thus, a mechanism may exist in microglia (following periods of ra pid division) that pref erentially recruits limiting amounts of telomerase to maintaining the shortest telomeres while allowing the longer telomeres to shorten (Ouellette et al., 2000). Telomerase activity may seem apparent when total protein is isolated and measured following in vitro culturing, yet the enzyme could be inhibited by a repressor molecule while in vitro or in vivo which may or may not be present within the total protein pool during anal ysis. Additional in vitro or in vivo molecules may also play critical roles in regulating telomerase activity. Thus, the pattern of telomerase ac tivity, as determined by in vitro total protein analysis, may not correlate precisely with or imply telomere maintenance in vitro or in vivo (Ouellette et al., 1999). Control microglia are already senescen t before day 32 while GM-CSF-stimulated cells are at or near senescence on the same day, suggesting that GM-CSF treatment may result in a slightly increased cel l life span due to the absence of critically-short telomeres.

PAGE 58

45 Our results suggest that the s lightly increased life span of GM-CSF-treated microglia may be due to their enhanced ability to maintain the shortest telomeres (by telomerase recruitment), which may result in their delayed entry into senescence, compared to controls. Perhaps during periods of rapi d division, microglia are able to more proficiently recruit telomerase to the shortest telomeres in an attempt to enable cell division to occur for a longer pe riod of time prior to entry into senescence. Therefore, the replicative capacity of rapidl y-dividing microglia appears to be greater compared to that of controls. Microglia were considered senescent by day 32 in control and GM-CSFtreated groups since they both exhibited decreased proliferat ion, telomere shortening, and altered phenotypes (relative to non-senescent dividing micr oglia). Despite continuous GM-CSF-stimulation, microglia were unable to maintain a high cell division rate, and most of the cells had sloughed off from the culture dish by day 32, indicating that they were no longer viable. Indeed, the yield of DNA was much lower, such that the entire DNA sample was used for Southern blot anal ysis for GM-CSF-treated cells on day 32, whereas only a fraction of the to tal DNA collected was used for an alysis in all other days. Microglia grown to near-confluence in various -sized culture dishes exhibit increased telomere attrition with increasing available culture area. Telomere loss likely occurs in these cultures due to additiona l cell divisions in microglia that are necessary to reach near-confluence in the larger-sized culture dishes. There were 4 times as many microglia plated in the 175 cm 2 flask compared to the 9.5 and 21 cm 2 dishes. However, the 175 cm 2 flask had a culture area that was over 18 times greater than the 9.5 cm 2 dish and over 8 times greater than the 21 cm 2 dish. Thus, it took approximately 4.5 more divisions per cell to reach near-confluence in the 175 cm 2 flask compared to the 9.5 cm 2 flask. These

PAGE 59

46 additional divisions likely account for the decrea se in the shortest telomeres evident in microglia cultured in the 175 cm 2 flask compared to the other smaller-sized culture dishes. Importantly, the s hortest telomeres in microgl ia cultured in the 175 cm 2 flask are up to 3.2 kb shorter, representing an 8.4% decrease, than thos e found in microglia cultured in the two smaller-sized dishes. These data demonstrate that the rate of erosion of the shortest telomeres in microglia is di rectly proportional to the number of divisions that a cell undergoes. Analysis of microglia telomeres by FISH re vealed that substantial telomere length variation occurs both between and within indivi dual chromosomes. This indicates that a heterogeneous population of telomeres exist within each microglial cell and among sister chromatids (Bekaert et al., 2002), and furthe r suggests that only a few individual chromosomes likely reach a critical length fi rst (thereby triggering cell senescence), assuming uniform telomere shortening with age. The telomeres that are the first to reach a critical length are likely the first to induce SAGE and TPE on these chromosomes. Thus, genes (especially those nearest the te lomeres) may be up-regulated with age in microglia due to TPE (Wright and Shay, 1992). Why does so much telomere length variation exist, especially within the same chromosome? Telomeres on individual chromosomes may be longer than those on ot her chromosomes in order to keep gene regulation and expression constant for that pa rticular area of the chromosome (especially genes nearest the telomere), or to prevent SAGE and TPE from occurring when telomeres shorten sufficiently. It would be interesting to determine which genes are located nearest the telomere on chromosomes which reach a crit ical length first, since this may provide clues as to the genetic changes that occur in mi croglia as they age. Furthermore, this may

PAGE 60

47 suggest an approach (e.g., telomerase over-e xpression) that would prevent TPE-induced expression of detrimental genes on chromo somes possessing very short telomeres. Future studies will permit quantitative FISH (Q -FISH) telomere length measurements, in which telomere lengths will be quantitated, co mpared to a standard curve of fluorescence intensity of plasmids containing known telomeric repeats (Poon et al., 1999). Since subtelomeric regions will not be bound by probe (as is the case for Southern blot), this will allow a more precise measurement of telomere length. Q-FISH will allow a determination of how much interand intra-chromosomal telomere length variation occurs in microglia and astrocytes with ag e, will help identify which chromosomes are the first to reach a critical length and trig ger senescence, and if certain chromosomes are more susceptible to telomere attrition. Telomeres in astrocytes shorten dramatically during the first few days of in vitro growth, which is consistent with their rapid division during this time to reach confluence. A fluctuating pattern of telomere maintenance and attrition follows, which is accompanied by transient telomerase up-re gulation during periods of telomere maintenance. This fluctuating cycle conti nues as the cells are maintained through day 32 without passaging, but it is unknown how long this cycle would continue. In astrocytes, senescence does not occur during these 32 days in culture. This shows that, at least in vitro astrocytes have a longer life span than microglia. Perhaps this is because astrocytes have longer telomeres overall than microglia, and/or because telomera se may be inhibited in astrocytes while in vivo but have high activity (due to lack of inhibition) when these cells are cultured. Small variations exist in telomere lengths between figures 5 and 6, both of which examine non-passaged astrocytes. This can be attributed to the numerous

PAGE 61

48 experimental parameters that can contribute to the slight variability present from one experiment to another due to the following: different exposure times (ranging from 1 to 2 hours) during development of the Southern bl ots, different hybridization times (ranging from 17 to 19 hours) that the membrane is in contact with the telomere probe, and especially since different cultures of cells ar e used for each experiment (each culture has cells with different rates of mitosis and varying concentr ations (about 1.5 rat pup brains per 175cm 2 flask) of whole brain cultures plated initially, which would lead to different division rates of both astrocytes and microglia prior to their collection, and could therefore account for th is variability). In passaged astrocytes, mean telomere le ngth remains nearly constant while the longest, and especially the shor test, telomeres fluctuate over time. This suggests that transient expression of low levels of telomera se may be occurring such that telomerase is preferentially recruited to the shortest telomeres, while mean telomere length remains unchanged (Ouellette et al., 2000) Thus, telomerase may be up-regulated (and therefore maintaining telomere length) during periods of rapid division, as when astrocytes are first passaged. Cultured astrocytes replicate cont inuously when passaged, compared to when they are plated in 9.5 cm 2 dishes and not passaged. Howe ver, it was noted that with continual passaging, a slowing of cell division occurred such that it took an increasingly longer amount of time to reach confluence with successive passages. Once astrocytes reach confluence in the 9.5 cm 2 dish, little replication is like ly to occur thereafter. Thus, non-passaged astrocytes undergo aging in the absence of cell division. However, when astrocytes are permitted to replicate conti nually (as when passaged), they maintain telomeres and even lengthen the shortest on es over time. As in GM-CSF-stimulated

PAGE 62

49 microglia, telomeres are initially lengthened when considerable replication is about to occur. Thus, given the opportunity to c ontinuously replicate, perhaps both microglia (when exposed to GM-CSF) and astrocytes (when passaged) up-regulate telomerase to compensate for telomere shortening that woul d otherwise occur. Future studies will examine telomere length and telomerase activity in microglia and astrocytes in vivo

PAGE 63

CHAPTER 3 TELOMERES SHORTEN WITH AGE IN RAT CEREBELLUM AND CORTEX IN VIVO Introduction Recent data from our laboratory (Flana ry and Streit, 2004) has shown that telomeres shorten with time in cultured microglia, suggesting that these brain cells are subject to replicative senescence in vivo Thus, we decided to determine how telomere length and telomerase activity ch ange in the rat brain with ag ing. Our results show that telomere erosion occurs in vivo in the rat cerebellum and co rtex with age in the presence of low levels of steadily in creasing telomerase activity. Materials and Methods Collection of Rat Cerebellum and Cortex Tissues Sprague-Dawley rats were maintained at 22 C in a controlled 12 hour light/dark cycle and provided food and water ad libitum Animals were euthanized by exsanguination using transcardiac perfusion with phosphate-buffere d saline under deep anesthesia with sodium pe ntobarbital (50 mg/kg body weight). This method of euthanasia is consistent with the recomme ndations of the Panel on Euthanasia of the American Veterinary Medical Association. Following perfusion, the cerebellum and cortex were dissected out and frozen prior to DNA and protein isolation. Determination of Telomere Length Telomere length was measured as described in Chapter 1. 50

PAGE 64

51 Determination of Telomerase Activity Telomerase activity was measured as described in Chapter 1. Results Telomeres Shorten With Age in Rat Brain in vivo Both rat cerebellum and cortex tis sue exhibit telomere shortening in vivo from day 21 to 152 (Fig. 3-1, 3-2). The cortex always had shorter telomere s (i.e., longest, mean, and shortest) than the cerebellum, except on da y 152 in the longest telomeres (Fig. 3-3). In cerebellum, the longest te lomeres shortened the most rapidly (loss of 5.6 kb, or 26%), followed by the mean telomeres (loss of 2.3 kb, or 17%), and the shortest telomeres (loss of 1.6 kb, or 18%) from day 21 to 152. In the cortex, the mean telomeres shortened most rapidly (loss of 2.9 kb, or 22%), followed by the longest telomeres (loss of 2.2 kb, or 12%), and the shortest telomeres (loss of 1.0 kb, or 13%) from day 21 to 152. In both cerebellum and cortex tissues, th e shortest telomeres were the slowest to shorten with age in vivo (Fig. 3-3). Telomerase Activity in Rat Cerebellum and Cortex Telomerase activity was measured in both cerebellum and cortex on various days. With increasing age in vivo telomerase activity steadily in creased in both tissues from day 21 to 182, with the cerebellum exhibiting the highest activity at all time points examined (Fig. 3-4, 3-5). From day 21 to 182, telomerase activity increa sed 28% in the cerebellum and 11% in the cortex. Further anal ysis of telomerase activity in cerebellum and cortex from day 21 to 35 revealed that ac tivity increased slightly in all samples with age in vivo except for one cerebellum sample from day 28 to 35 (Fig. 3-6, 3-7). Overall, telomerase activity increased with age in vivo from day 21 to 35 in both cerebellum and cortex, with the cerebellum exhibiting the highest activity at all time points measured.

PAGE 65

52 From day 21 to 35, telomerase activity incr eased 12% in the cerebellum and 11% in the cortex (Fig. 3-8). Figure 3-1. Southern blot analysis for meas urement of telomere restriction fragment (TRF) length in rat brain tissue. Genomic DNA was isolated on the indicated days (day 152 is approximately 5 months of age) from cerebellum and cortex tissues of two different rats (A and B).

PAGE 66

53 4 6 8 10 12 14 16 18 20 22 24TRF Length (kb) Longest Mean Shortest Longest 22.220.219.318.215.016.316.6 Mean 14.013.012.813.811.211.110.4 Shortest 8.2 9.3 8.0 8.0 7.4 6.8 7.0 Cereb A 21Cereb B 21Cortex A 21Cortex B 21Cereb A 152Cereb B 152Cortex 152 Figure 3-2. TRF length distribut ion in rat cerebellum and co rtex samples on days 21 and 152. 0 2 4 6 8 10 12 14 16 18 20 22 24Average TRF Length (kb) Longest Mean Shortest Longest 21.2 15.6 18.8 16.6 Mean 13.5 11.2 13.3 10.4 Shortest 8 77 18 07 0 Cereb 21Cereb 152Cortex 21Cortex 152 Figure 3-3. Average TRF length in rat cerebe llum and cortex tissues on days 21 and 152. Overall, the cerebellum has longer telome res than the cortex from day 21 to 152 in all instances (except in longe st telomeres on day 152). Telomere shortening occurs with age in vivo in both rat cerebellum and cortex. Although only two animals were analyzed for each time point (except cortex day 152), no overlap of erro r bars exists between cereb ellum and cortex in any instances.

PAGE 67

54 Figure 3-4. TRAP analysis for telomerase act ivity in rat brain tissue (days 21 to 182). Total protein was isolated on the indicated days (days 152 and 182 are approximately 5 and 6 months of age, respectively) from cerebellum and cortex tissues. Neg = telomerase-neg ative control (i.e., cortex A day 21 RNAse-treated extract). Pos = telomerase-positive control.

PAGE 68

55 24 26 28 30 32 34 36 38 020406080100120140160180200Time (days)Average Telomerase Activit y Cortex A Cereb A Figure 3-5. Quantitation of telomerase activit y (arbitrary units) in rat cerebellum and cortex tissues (days 21 to 182). Telome rase activity gradually increases with age in vivo in both cerebellum and cortex in all instances. Figure 3-6. TRAP analysis for telomerase act ivity in rat brain ti ssue (days 21 to 35). Total protein was isolated on the indicat ed days from cerebellum and cortex tissues of two different rats (B and C).

PAGE 69

56 31 32 33 34 35 36 37 38 39 202224262830323436Time (Postnatal Day)Average Telomerase Activity Cortex B Cereb B Cortex C Cereb C Figure 3-7. Quantitation of telomerase activit y (arbitrary units) in rat cerebellum and cortex tissues (days 21 to 35). Telome rase activity gradually increases with age in vivo in both cerebellum and cortex in all instances (except cerebellum B from day 28 to 35). 31 32 33 34 35 36 37 38 202224262830323436Time (days)Average Telomerase Activity Cortex Cereb Figure 3-8. Overall, the cerebellum exhibits higher telomerase activity than the cortex from day 21 to 35.

PAGE 70

57 Discussion This study, which is the first to report on both telomere length and telomerase activity in a region-specific manner in rat brai n, shows that telomere shortening occurs in both rat cerebellum and cortex with increasing age in vivo The telomere shortening is accompanied by low levels of steadily increasing telomerase activity, which is highest in the cerebellum in all instances. Our current findings on telomere length in brain tissues are in apparent conflict with a recently published study (C herif et al., 2003). These authors report that while telomeres shorten with age in rat kidney, liver, lung, and pancreas, no telomere shortening occurs in rat brai n tissue with age (from postnat al day 21 to 15 months of age). However, their data does indicate that both the longest and s hortest telomeres do shorten in rat brain with age, but not significantly. In additio n, the analyses in the study (Cherif et al., 2003) were based on unspecified brain areas, while our current findings are in defined brain regions and showed corresponding results (i.e., telo meres shorten in rat brain with age). Moreover, it appears that the animals used in the study (Cherif et al., 2003) were not perfused prior to tissue collection, and thus it is likely that the brain tissue used for telomere length analysis represente d a mixture of brain a nd blood tissue. Blood contamination of brain tissue could have obscured any brain-specific telomere shortening. On the other hand, there are tw o independent studies which have reported that telomeres do undergo shortening in brain tissue with age. One study performed in mice demonstrated that telomere shortening does occur in spleen and brain tissue, but not in liver, testes, or kidney with age from zero to nine months (Prowse and Greider, 1995). Another study also found that telomeres shorte n in mice brain tissue with age from one to 24 months (Coviello-McLaughlin and Prowse, 1997). These three studi es (Cherif et al.,

PAGE 71

58 2003; Prowse and Greider et al., 1995; Covi ello-McLaughlin and Prowse, 1997) are the only ones known, at the time of this writing, th at have examined telomere length in brain tissue of rodents with age. Clearly, more wo rk will be required to resolve any conflicting results that may exist now and to reach a satisfactory conclusion a bout telomere erosion in the CNS. The rate of telomere loss of the longest telomeres in the cerebellum from day 21 to 152 is over two times greater than that in the cortex. This su ggests that more cell division occurs in the cerebellum during this time frame in vivo, compared to the cortex. Our current study has revealed a difference in telomere lengths between the cerebellum and cortex by showing that the longest, mean, and the shortest telomeres in the cortex are consistently shorter than those found in the cerebellum in all but one instance. This observation may, at first, seem unexpected si nce cerebellar, unlike cortical, histogenesis is characterized by the existen ce of a secondary proliferative zone during late stages of cerebellar development (Steward, 2000), and one might therefore expect to see shorter telomeres in the cerebellum. However, our da ta also show higher telomerase activity in the cerebellum, compared to the cortex, which thereby may account for the presence of longer telomeres (despite cell division in the secondary prolifer ative zone) in the cerebellum compared to the cortex. Few studies have examined telomerase activity in the brain. One study (Klapper et al ., 2001) reported that telomera se activity is high in mouse cortex during embryonic development, but sharply decreases during postnatal development up to three months of age. Likewise, another study (Fu et al., 2000) reported that telomerase activity is high in rodent neurons during embryonic and early postnatal development, but then subsequently decreases. Telomerase activity is also high

PAGE 72

59 in rat oligodendrocyte precurs or cells, but declines duri ng their differentiation into oligodendrocytes (Caporaso and Chao, 2001). Our current results indicate that telomerase activity is lowest on day 21 and st eadily increases, albeit slightly, up to six months of age. In comparison to previous research in this area (K lapper et al., 2001), our results may or may not coincide. The slight increase in telomerase activity reported in our study from day 21 to 6 months of age may similar to the same low/decreased levels as those found in previous studies (Kla pper et al., 2001) (after early postnatal development) if sample sets from both studies were compared on the same gel. It is nearly impossible to make an accurate comparison between different TRAP gels (especially between differe nt laboratories), since ma ny factors (e.g., annealing temperature, number of PCR cycles, prot ein and primer concentration used, etc.) influence ladder formation, band intensity, and subsequent telomerase activity levels (which are expressed in arbitrar y units, or as a percentage of the highest levels evident). The longest telomeres (in cerebellum) a nd the mean telomeres (in cortex) undergo the highest rate of telomere loss with age in vivo However, in addition to the longest telomeres in the cortex, the shortest telomere s in both tissue types exhibit the slowest rate of attrition with age. Since telomerase activit y slightly increases in both tissue types with age, this suggests that limiting amounts of telomerase may be preferentially recruited to the shortest telomeres, while allowing the l ongest telomeres to shor ten (Ouellette et al., 2000). This data also suggest s that the amount of telomerase present may not be enough to sufficiently compensate for the rate of telomere loss that occurs in the shortest telomeres with age, since telomeres in bot h tissue types still s horten with age.

PAGE 73

60 There is high mitotic activity of differe nt, primarily glial, cell types in the developing rodent brain during the neonatal/postnatal peri od up to about postnatal day 14. On postnatal day 21, most mitotic activity has ended a nd the CNS is almost fully matured. At this time, microglia are probabl y the only adult cell type remaining in the postnatal CNS that undergo any appreciable ce ll division, and they retain their mitotic ability throughout adult life. Since our analys es began on day 21, we are inclined to think that most of the telomere erosion subsequent to this day that occurs in both cerebellum and cortex may be largely attributable to ce ll division of microglia. Of course, we cannot exclude the possibility that so me of the observed telomere a ttrition may also be also be contributed by neural stem cells in the subventricular zone s. Future studies employing tissue from the hippocampus could further i lluminate this issue, since neurogenesis occurs in the hippocampus throughout adult lif e, but declines with aging (Kuhn et al., 1996). In addition, some of the observed telomere attrition could also be attributable to mechanisms not related to cell division, such as oxidative stress (von Zglinicki, 2002), in postmitotic cells. Future studies in our laboratory will examine telomere length and telomerase activity in vivo in rat cerebellum and cortex over a longer time period (up to several years) as well as in acutely isol ated microglial cells from young and aged adult rats.

PAGE 74

CHAPTER 4 AXOTOMY INCREASES TELOMERE LENGTH, TELOMERASE ACTIVITY AND PROTEIN IN AXOTOMY-ACTIVATED MICROGLIA Introduction Previous research in our la boratory has demonstrated th at telomere shortening and senescence occurs in cultured rat microglia following periods of prolonged and sustained mitotic activity induced by GM-CSF (Flanary and Streit, 2004). In contrast, microglia in vivo undergo short proliferative bursts soon after an acute injury has occurred. This, together with the fact that microglia produ ce growth factors and cy tokines after injury, suggests that mitosis affords a mechanism to provide greater numbers of microglial cells and thus greater trophic support during times of stress (Streit et al ., 2000; Streit, 2002a). However, the mitotic potential of microglia also suggests that these cells have limited cellular life-spans and thus may rely on proliferation a nd self-renewal to replace senescent cells. In order to determine whether neuronal inju ry-induced microglial proliferation within a well-defined region of the CNS results in microglial telomere shortening or cellular senescence, we deci ded to measure parameters indicative of telomeric maintenance, such as telomere length and telomerase activity. Materials and Methods Rat Facial Nerve Axotomy Adult Sprague-Dawley rats of both genders were housed at 22 C in a controlled 12 hour light/dark cycle and provided food and water ad libitum Animals were anesthetized using isofluorane, and the right facial nerve was expos ed and transected at the 61

PAGE 75

62 stylomastoid foramen. Failure to move whiske rs on the right side of the face following recovery from anesthesia was used to verify the success of the axotomy. Animals were euthanized at 1, 4, 5, 7, and 10 days post-ax otomy by exsanguination using transcardiac perfusion with phosphate-buffered saline (PBS) under deep anesthesia with sodium pentobarbital (50 mg/kg body weight). This met hod of euthanasia is consistent with the recommendations of the Panel on Euthanasia of the American Veterinary Medical Association. Following pe rfusion, the entire brain was removed, snap-frozen 2methylbutane cooled by liquid ni trogen, and stored at -80 C. Individual facial nuclei of both axotomized (i.e., right) and control (i.e ., left) sides were mi cro-dissected using a cryostat and stored at -80 C prior to DNA and protein isolation. FACS-Isolation of Rat Microglia fr om Micro-dissected Facial Nuclei Animals were exsanguinated under deep sodium pentobarbital anesthesia (50 mg/kg body weight) using ice-cold PBS. The brains were removed and placed in PBS on ice. Micro-dissection of facial nuclei from the chilled brains was performed immediately after removal. Nuclei (approximately 7 mg wet weight each) from both axotomized and non-axotomized sides were collected from six animals at 5 days post-axotomy. Six nuclei from each side were pooled and pro cessed according to established isolation protocols (Carson et al., 1998). Fluorescence-activated cell so rting (FACS) analysis was performed using a FACSVantage SE cell sorter and CellQuest software (BD Biosciences/Becton Dickinson, San Jose, CA). Monoclonal antibodies used to isolate microglia during FACS analysis were FITC-conjugated antirat CD45 (leukocyte common antigen), and PE-conjugated an ti-rat CD11b/c (CR3 complement) (BD

PAGE 76

63 Biosciences/Pharmingen, San Diego, CA), with microglia being identified as the CD11b/c high and CD45 low cell population (F ord et al., 1995; Sedgwick et al., 1991). Determination of Telomere Length Telomere length was measured as described in Chapter 1. Determination of Telomerase Activity Telomerase activity was measured as described in Chapter 1. Telomerase Western Blot Analysis Telomerase protein quantity was measured using Western blot analysis and chemiluminescent detection. Total protein was isolated as described above, and 50 g from each sample, and a telomerase-positive control (rat glioblastoma cell line RG-2), was separated on an 8% SDS-PAGE gel at 25 mA for 1 hour. Protein was transferred from the gel to an Immobilon PVDF (polyvinylidene fluoride) membrane (Millipore, Billerica, MA) using semi-dry transfer at 5 volts for 1 hour. The membrane was blocked in 5% milk solution in TBST (tris-buffere d saline with 0.1% Tween-20) for one hour at room temperature with shaking, and then incubated overnight in a primary antibody (rabbit polyclonal anti-telomerase antibody: EST-21A) (Alpha Diagnostics, San Antonio, TX) at a 1:250 dilution (4 g/mL) in 5% milk solution in TBST at 4 C with gentle agitation. A second primary antibody (r abbit polyclonal anti-telomerase antibody: NB 100-141) (Novus Biologicals, Littleton, CO) was also used in parallel experiments with the Alpha Diagnostics antibody. Both an tibodies yielded similar banding patterns, however, we found that the Novus Biologica ls antibody gave very high background, and thus used the antibody from Alpha Diagnostics Following washes in TBST for 4 x 15 min. at room temperature with fast shaking (120 rpm), a secondary antibody (horseradish peroxidase (HRP)-conjugated anti-rabbit) was applied for one hour at room temperature

PAGE 77

64 with gentle agitation. Following washes in TBST for 4 x 15 min. at room temperature with fast shaking (120 rpm), chemiluminescent detection was accomplished by incubating the membrane in the HRP-metabolizing substrate ECL (enhanced chemiluminescence) (Amersham Biosciences, Piscataway, NJ) for one minute at room temperature. Following exposure of the memb rane to X-omat AR Film (Eastman Kodak Company, Rochester, NY) in an autoradiograp hy cassette for 1 to 10 min., the film was developed using a Konica SRX-101A auto matic film processor (Konica Minolta, Mahwah, NJ). A digital image of the auto radiograph was generated by scanning it using a GS-710 calibrated imaging densitometer (B ioRad, Hercules, CA). Quantitation of telomerase protein quantity was performed using the densitometry computer program Quantity One (version 4.3.1) (BioRad, Hercules, CA). Following development, the membrane was re-probed with an anti-GAPDH (glyceraldehyde-3-phosphate dehydrogenase) antibody as a lo ading control. The band corresponding to telomerase protein (catalytic portion) appeared at approximately 60 kDa, while GAPDH appeared at approximately 40 kDa. Normalized telome rase protein levels were determined by comparing the average densitometric values of identical samples ran on different gels in order to make an accurate comparison of protein levels between all samples run on multiple gels. Histochemistry Animals were perfused with 4% paraformaldehyde at 3 days post-axotomy. Dissected tissues were post-fixed for severa l days, then rinsed and stored in PBS. Following paraffin-embedding, serial sections (7.0 m thick) were cut on a microtome (model 2040, Reichert-Jung, Buffalo, NY), m ounted onto gelatin-coated slides, and

PAGE 78

65 allowed to dry. Deparaffinization and rehydra tion were performed by soaking the slides in xylenes (2 x 15 min.), followed by passa ge through descending ethanols, including soaking in 70% ethanol overnight. Lectin histochemistry for the localization of microglial cells, using the Griffonia simplicifolia B4 isolectin coupled to horseradish peroxidase (i.e., GSI B4-HRP) (catalog # L5391, Sigma-Aldrich, St. Louis, MO), was performed as described by Streit (1990). Following development with the peroxidase substrate diaminobenzadine (DAB), slides we re coverslipped with Permount. Selected sections were counterstained w ith cresyl violet. Slides we re photographed using a Zeiss Axioskop 2 plus microscope with a RT color Spot camera (model 2.2.1, Diagnostic Instruments, Inc., Sterling Heights, MI). Photographs were edited using Spot software (version 3.4.5) (Diagnostic Instruments, Inc., Sterling Heights, MI). Statistical Analysis of Data Data was analyzed using the software program InStat version 3.06 (GraphPad, San Diego, CA). In order to determine whether a statistically signifi cant difference existed between treatment groups, analysis of va riance (ANOVA) was performed using the Tukey-Kramer multiple comparison test for TR AP assays, while an unpaired t-test was performed while for Western bl ot experiments. A p value of >0.05 was considered nonsignificant. Results Increase in Microglia Surrounding Axotomized Facial Nuclei Histological examination of sections conf irmed prior descript ions of microglial activation in the axotomized facial nucleus (F ig. 4-1). On the axotomized side, increased numbers of glial nuclei were a pparent and increased lectin staining confirmed that this was due to greater numbers of microglia. Th e activated microglia often were gathered

PAGE 79

66 around axotomized motor neurons encircling them with their processes. Nuclear staining with cresyl violet revealed the presence of mitotic figures showing active microglial proliferation. Increase in Telomere Length in Axotomized Facial Nuclei Southern blotting was used to perform te lomere length analysis in pooled microdissected rat facial nuclei (Fig. 4-2), with quantitation shown in Fig. 4-3. Telomere length in axotomized facial nuclei (4 days post-axotomy) increased in the longest (by 21%), mean (by 9%), and shortest (by 10%) te lomeres, compared to the control facial nuclei. This indicates that telomeres le ngthen in whole facial nuclei tissue following axotomy.

PAGE 80

67 Figure 4-1. Micrographs of a xotomized (A) and control (B) facial nucleus on day 3 postaxotomy stained with GSI-B4 lectin to identify microglia. Note the mitotic figure (black arrow, higher magnifica tion shown in inset), and increased microglial cell numbers on the axotomized side, particularly around injured motor neurons (red arrows). Scale bar = 25 m at 40X magnification (A and B), and 10 m at 100X magnification (inset).

PAGE 81

68 Figure 4-2. Southern blot analysis for measur ement of telomere length in facial nuclei.

PAGE 82

69 Figure 4-3. Densitometric quantitation of telomere length in facial nuclei. Genomic DNA was isolated from 10 pooled axotom ized (A) and unlesioned control (C) facial nuclei on day 4 post-axotomy a nd probed with a digoxigenin-labelled telomere specific oligonucleotide (TTAGGG 3 ). Densitometric quantitation revealed an increase in telomere le ngth in axotomized facial nuclei. Increase in Telomerase Activity in Axotomized Facial Nuclei To test whether the observed increases in telomere length were attributable to similar increases in telomerase activity, TRAP analysis was used for determination of telomerase activity from day 1 to 10 post-axotom y in individual micro-dissected rat facial nuclei (Fig. 4-4). Normalized telomerase activity (i.e., of 3 separate TRAP gels with different samples normalized relative to each other) was significantly increased in the axotomized facial nuclei (compared to control nuclei) on days 1, 4, 7, and 10 by 8.96%, 68.74%, 104.05%, and 48.25%, respectiv ely (Fig. 4-5). In a non-axotomized animal, no significant difference in telomerase activity existed between the right and left facial

PAGE 83

70 nuclei (Fig. 4-6). These data suggest that one or more different cell types up-regulate telomerase activity following axotomy. Figure 4-4. Representative TR AP analysis image used for measurement of telomerase activity in facial nuclei. Neg = telomerase-negative control; Pos = telomerase-positive control; IC = internal control.

PAGE 84

71 Figure 4-5. Densitometric quantitation of telomerase activity in facial nuclei. Total protein was isolated from axotomized (A ) or control (C) facial nuclei (day 4 post-axotomy, n = 4). Densitometric quantitation revealed an increase in mean ( SEM) normalized telomerase act ivity (arbitrary units) in axotomized facial nuclei at all timepoints, compared to controls (B). The percent increase in normalized telomerase activity in axotomized facial nuclei (compared to controls) following facial nerve axotom y increased the most from day 1 to 4, and peaked on approximately day 7. Day 0 represents a non-axotomized animal. Telomerase activity was sign ificantly increased on day 4 (p<0.01), day 7 (p<0.001), and day 10 (p<0.05) compared to day 1, while day 7 was higher than day 10 (p<0.01). Total number of animals analyzed on each day was: day 1 (5), day 4 (6), day 7 (6), day 10 (5).

PAGE 85

72 Figure 4-6. Densitometric quantitation of telome rase activity in unoperated facial nuclei. Absence of variation (p>0.05) in mean ( SD) telomerase activity (arbitrary units) between left and right non-axotomized facial nuclei from nave, unoperated animals (C). Five independent samples were analyzed for both the left and right facial nuclei. Telomerase activity in the right facial nuclei was 1.42% higher than in left f acial nuclei, but was not statistically significant. Increase in Telomerase Protein Quanti ty in Axotomized Facial Nuclei In order to determine if the increases evid ent in telomerase activity were due to an increased protein expression, We stern blotting was used for determination of telomerase protein quantity in individual micro-dissected rat facial nuclei (Fig. 4-7). Quantitation of telomerase protein in rat facial nuclei is shown in Fig. 4-8. Normalized telomerase protein quantity (i.e., of 3 separate Western blot gels with different samples normalized relative to each other and to GAPDH) was higher in axotomi zed facial nuclei (compared to control nuclei) on day 1 (by 27.68%), 4 (by 45.13%), 7 (by 37.01%%), and 10 (by

PAGE 86

73 103.16%) post-axotomy. This suggests that one or more different cell types up-regulate telomerase protein following axotomy. Figure 4-7. Western blot image used for meas urement of telomerase protein quantity in facial nuclei. Total protein was isolat ed from axotomized (A) or control (C) facial nuclei on either 1, 4, 7, or 10 days post-axotomy (A). Pos = telomerasepositive control (RG-2 glioma cells). Figure 4-8. Densitometric quantitation of telomerase protein in f acial nuclei. An increase in mean ( SEM) normalized telomerase protein (arbitrary units) in axotomized facial nuclei at all timepoin ts was observed, compared to controls (B). The percent increas e in normalized telomerase protein in axotomized facial nuclei (compared to controls) following facial nerve axotomy increased the most from day 1 to 4, and slowly declined from day 4 to 10. Telomerase protein was significantly increased (p<0.03) on day 4 compared to day 1 in the axotomized side; all other timepoints are non-significant (p>0.05). Axot = axotomized facial nuclei; Con = control facial nuclei.

PAGE 87

74 FACS-Isolation of Microglia from Facial Nuclei To determine if microglia were the cell type responsible for the increases in telomerase activity/protein (and hence telomere length), microglia were isolated from six pooled facial nuclei using CD11b/c and CD45 an tibodies (Fig. 4-9). Since the samples contained mixed cell populations cells were initially sorted based on the R1 gate, which excluded cellular debris (near the lower-l eft corner) and other undesirable cells (e.g., doublet and triplet cells) present elsewhere within the plot (o ther than R1). Microglia were identified as the CD11b/c high and CD45 low cell population (i.e., the R2 gate). The parameters of the R1 and R2 gates were determined in previous experiments (Carson et al., 1998). A total of 1 ,334 microglia (0.46% gated) were isolated from the pooled axotomized facial nuclei (Fig. 4-9H), wh ereas only 669 total microglia (0.22% gated) were isolated from the pooled control facial nuclei (Fig. 4-9J), indicating that at least twice as many microglia were present in the axotomized facial nucleus compared to the control nucleus. Increase in Telomerase Activity in F ACS-Isolated Microglia From Axotomized Facial Nuclei TRAP analysis was used for determination of telomerase activity in FACS-isolated microglia from rat facial nuclei (Fig. 4-10). Quantitation of averag e telomerase activity in FACS-isolated microglia is shown in Fi g. 4-11. Telomerase activity increased by 254.71% in FACS-isolated microglia from axotom ized facial nuclei compared to control facial nuclei. This indicates that microglia within the axotomized facial nucleus exhibit a large increase in telomerase activity when analyzed separa tely (i.e., apart from whole tissue samples), suggesting that they are the cell type mainly responsible for the increases evident in telomerase activity/protein, and hence telomere length, following axotomy.

PAGE 88

75

PAGE 89

76 Figure 4-9. FACS-isolation of microglia from axotomized and control facial nuclei. Facial nuclei (axotomized or control) were pooled from rats (5 days postaxotomy) and subjected to FACS analysis. Antibodies to CD11b/c and CD45 were used to isolated microglia. A and B are the negative control (no antibodies used); C and D are the CD11b/c antibody only; E and F are the CD45 antibody only; G and H are the axotom ized facial nuclei; I and J are the control facial nuclei. Cells within R1 were selected and sorted via R2 gating to isolate microglia.

PAGE 90

77 Figure 4-10. TRAP analysis image used fo r measurement of telomerase activity in FACS-isolated facial nuclei. Figure 4-11. Densitometric quantitation of te lomerase activity in FACS-isolated facial nuclei. Total protein was isolated from FACS-isolated microglia from axotomized (A) and control (C) facial nuclei (day 5 post-axotomy, six independent animals, two separate expe riments). IC = internal control. Telomerase activity (arbitrary units) w ithin FACS-isolated microglia from axotomized facial nuclei on day 5 post-axotomy was increased by 254.71% compared to the control facial nuclei.

PAGE 91

78 Discussion We have shown that facial nerve axotomy results in a probable increase in telomere length and telomerase protein, and a definite increase in telomerase activity in the axotomized facial nucleus. A definite incr ease in telomerase activity also occurs in FACS-isolated activated microglia from the axotomized facial nucleus. These results support the hypothesis that maintenance and ex tension of telomere length occurs in activated microglia that accumulate in axotomi zed rat facial nuclei. Telomere extension is likely the result of the observed increases in telomera se protein and activity. Our findings suggest that dividing microglial cells of the CNS compensate for replicationinduced telomere shortening by up-regulating telomerase. A total of 10 facial nuclei were pooled for use in Southe rn blot analysis. This pooling of tissue samples was necessary due to their small size (approximately 7 mg wet weight each) and in order to load sufficient quantities of DNA in order to generate a signal during the Southern blot detection process. Axotomy cause d a trend towards an increase in telomere length (in the longest, mean, and shortest) in the axotomized facial nuclei, compared to the control facial nuclei. A concomitant increase in both telomerase activity and telomerase protein was also observed in axotomized facial nuclei. Since microglia are the only cells know n to divide in response to axotomy (Graeber et al., 1988), these findings suggest that microglia are utilizing telomerase to regulate telomere length in vivo during periods of high proliferation. We believe that this presumed increase in microglial telomere length in the axotomized facial nuclei compensates for the shortening of telomeres that would otherwise occur in the absence of such telomerase activity. Since FACS-isolated microglia from the axotomized facial nuclei exhibit an increase in telomerase activity, the increas e in telomere length evident within whole

PAGE 92

79 facial nuclei may be attribut able to the increased telomerase activity present within microglia. It remains possible that other cells types may be present which up-regulate expression, and hence activity, of telomerase following axotomy. Telomerase activity increased the most from day 1 to 4, and p eaked on day 7 post-axotomy, representing a maximal increase in activity of 104.05%. In view of proliferation data from previous studies (Graeber et al., 1988; Kreutzberg, 1966; Streit and Kreutzberg, 1988; Svensson et al., 1994), which showed a peak in microglia l proliferation on da y 3 post-axotomy, our findings suggest that the increase in telomera se activity evident within the first 4 days following axotomy is to prepare microglia for their concomitant proliferative burst (occurring on days 2 to 4 post-axotomy). Telo merase activity conti nued to rise from day 4 to 7 as well, indicating the presence of enzymatic activity to lengthen telomeres, if needed. From 7 to 10 days post-axotomy, te lomerase activity dec lined, possibly since additional proliferation does not occur dur ing this time period (G raeber et al., 1988; Kreutzberg, 1996; Streit and Kr eutzberg, 1988; Svensson et al., 1994), and thus telomere length would likely remain stable, and hence te lomerase activity would not be required. Analysis of telomerase activity in nave (i.e., non-axotomized) animals showed no difference in telomerase activity between the tw o sides, as expected. The processivity of the telomerase enzyme (Greider, 1991) (i.e., the quantity of telomeric repeats processed by the enzyme, as determined by banding patter n on gel), in both whol e facial nuclei and in microglia FACS-isolated from facial nuclei, correlated well with measured telomerase activity. Lanes possessing an in tense initial band (immediately above the internal control band) always contained a large quantity/intensity of small telomerase products, as well as a small quantity/intensity of larger products and are indicative of samples with low

PAGE 93

80 enzyme processivity (i.e., cont rol side of injured facial nuc leus). The presence of the intense initial band indicates that, most of the time, telomerase added only a single hexanucleotide telomeric repeat onto the end of the TS primer before dissociating. Thus, most of the measured telomerase activity wa s contributed by the presence of the first few initial bands. In these lanes, telomerase apparently is only able to generate small products and lacks the ability to create larger products, whic h are clearly present in lanes with high telomerase processi vity (i.e., axotomized side of injured facial nucleus). There was a two-fold increase in the numb er of microglia isolated by FACS from the axotomized facial nuclei (i.e., 1,334) compar ed to the control faci al nuclei (i.e., 669), which was expected since microglia are know n to proliferate in response to axotomy (Graeber et al., 1988; Kreutzbe rg, 1996). Since TRAP analysis is a PCR-based assay, it enabled the measurement of te lomerase activity from such small numbers of cells. FACS-isolated microglia from facial nuclei e xhibited an increase in telomerase activity (by 255%) compared to the control facial nu clei, and thus the increase in telomerase activity evident within whole facial nuclei tis sue samples is most likely attributable to microglia. Importantly, telomerase activity in FACS-isolated microglia (255% higher in axotomized side on day 5 post-axotomy) is 270% higher compared to that in whole facial nuclei tissue samples (69% higher in axotom ized side on day 4 post-axotomy). Thus, when telomerase activity is measured in whol e tissue samples, cells other than microglia which may have low telomerase activity, likely account for th e overall decrease in enzymatic activity. Few studies have examined telomere length or telomerase activity in the brain, and most published research in this area has focused on CNS tumors and neural precursor

PAGE 94

81 cells. We have previously shown that telomere shortening occurs in cultured rat microglia in vitro (Flanary and Streit, 2004), and in rat cerebellum and cortex in vivo in the presence of low levels of telomerase activity (Flanary and Streit, 2003). One study reported that telomerase activity is high in mouse cortex during embryonic development, but sharply decreases during postn atal development up to three months of age (Klapper et al., 2001). Telomerase activity is also high in rat oligodendrocyt e precursor cells, but declines during their differentiation into ma ture oligodendrocytes (Caporaso and Chao, 2001). Human neural precursor cells express low levels of te lomerase at early passages, with levels declining to undetectable levels at later passages (greater than 20 population doublings). In contrast, rodent neural precursor cells express high levels of telomerase at both early and late passages (Ostenfeld et al., 2000). Telomerase has been found to be expressed in all brain regions shortly after birth, but becomes restricted to neural stem cells within the subventricular zone and olfactory bulb in the adult mouse brain (Caporaso et al., 2003). Likewi se, another study reported that telomerase activity is high in rodent neurons during embryonic and early postnatal development, but decreases subsequently (Fu et al., 2002). The latter st udy also found that suppression of telomerase expression in cultured embryonic hippocampal ne urons increased their vulnerability to apoptosis and excitotoxicity, suggesting that telomerase may play roles other than telomere maintenance. Induction of telome rase in neurons has been found to exhibit neuroprotective properties in e xperimental animal models of neurodegenerative disorders (Mattson, 2000). Another study reported the induction of te lomerase mRNA in cortical neurons following ischemia (Kang et al., 2004). Thus, telomerase appears to play critical roles during embryonic development and fo llowing brain injury, and it may be

PAGE 95

82 neuroprotective in non-dividing neurons by pe rforming functions unrelated to telomere length maintenance, such as repair of te lomeres damaged by free radicals (von Zglinicki, 2002). The results from our current study in ra ts also suggest a link between telomerase and neuroprotection, in that increased te lomerase activity may prevent microglial senescence thereby ensuring sustained glial support of injured neurons. An acute increase in activated microglia in vivo following axotomy not onl y serves a beneficial role, but is also a crucial com ponent of the regenerative proce ss, since microglia divide in response to the injury, surround and ensheath injured motor neurons, and provide them with trophic support (Streit, 2002b). However, if multiple bouts of proliferation occurs (e.g., in response to repeated in jury), this may result in an accelerated rate of telomere shortening in microglia, which may hasten their entry into replicative senescence, and may thereby limit both the quantity and quality of glial support they are able to provide to neurons. Recently, astrocytes and microglia have been shown to express telomerase immunoreactivity in vivo following ischemic or kainite-i nduced brain injury (Baek et al., 2004; Fu et al., 2002). We al so performed immunohistochemistry for the detection of telomerase reverse transcriptase, but thes e experiments were unsuccessful in that no specific immunoreactivity was observed. The experiments were performed with two different anti-telomerase antibodies (ra bbit anti-telomerase antibody, NB 100-141, from Novus Biologicals, Littleton, CO; or rabb it anti-telomerase an tibody, EST-21A, from Alpha Diagnostics, San Antonio, TX), us ing both 4% paraformaldehyde and 10% formalin fixation with and without antigen retrieval in 0.01 M citrate buffer. We have shown by Western blot analysis that telomerase protein is increased in the axotomized facial nucleus compared to the co ntrol side. On day 10, there is over a 100%

PAGE 96

83 increase in telomerase activity, which corres ponds to, and is likely caused by, the higher rate of decline of telomerase protein in the control compared to the axotomized side from day 7 to 10 post-axotomy. It is unknown what causes telomerase protein to decrease at a slightly faster rate in the control side (relative to the axotomized side) from day 7 to 10 post-axotomy. On day 10, the quantity of telome rase protein is markedly higher than day 1 in the axotomized side, but is slightly lowe r in controls relative to day 1, suggesting that telomerase protein (and activity) are necessary in the axotomized side up to at least day 10 post-axotomy. The largest increase de tected was on day 4 post-axotomy (45.13% increase) (p<0.03), followed by day 7 (37.01% increase). This data corresponds well with the microglial proliferative burst after axotomy in that the largest increase in the production of telomerase protein is coincident with the maximal number of microglial cells present 4 days after the axotomy. The increase in telomerase protein also shows good temporal correspondence with a large increa se in telomerase activity suggesting that increased enzymatic activity may be due to th e observed increase in telomerase protein. Specifically, the increase in telomerase protein preceded the increase in telomerase activity, since the percent increase in pr otein peaked on day 4 (45.13% higher in axotomized facial nuclei), whereas ac tivity peaked on day 7 (104.05% higher in axotomized facial nuclei). This data suggest s that telomerase is translationally-regulated, at the least. Interestingly, the temporal pr ofiles of telomerase pr otein followed parallel patterns in the axotomized and control faci al nuclei. Both axotomized and unoperated control facial nuclei showed increases in telomerase protein from day 1 to 4, and a decrease from day 4 to 10. This sugg ests that unilatera l axotomy can produce

PAGE 97

84 contralateral effects, an interesting phenom enon that has been noted before but remains exceedingly difficult to define because of its subtlety and inconsistency. The findings from our previous studies in vitro (Flanary and Streit, 2004) provided an impetus to further characterize telomere length and telomerase act ivity in the facial nucleus following repeated axotomies to i nduce continuous mitogenic stimulation. We found that microglia exposed to conti nuous mitogenic stimulation with GM-CSF in vitro show a dramatic increase in telomere length during the first few days post-stimulation, followed by rapid telomere shortening and se nescence. In the current study, a similar situation is observed in that after micr oglia undergo a short burst in proliferation following axotomy, a resultant increase is observed in telomere length. Telomere shortening may become evident when microgl ia are subjected to multiple rounds of proliferative bursts induced by repeated axotom ies. These experiments will be the focus of future studies, and may support the hypothesis that repeated brain injury could lead to microglial senescence.

PAGE 98

CHAPTER 5 ALPHA-TOCOPHEROL (VITAMIN E) INDUCES RAPID, NON-SUSTAINED PROLIFERATION IN CULTURED RAT MICROGLIA Introduction Microglial Activation Activation of microglial cells is a critical component of the brains response to injury. It is particularly prominent after acute lesions when it occurs rapidly and is characterized, among other things, by a dram atic increase in microglial cell numbers (Graeber et al., 1988; Kreutz berg, 1996). Granulocyte m acrophage-colony stimulating factor (GM-CSF) has long been known to be an effective microglial mitogen in vitro (Giulian and Ingeman, 1988; Suzumura et al., 19 90), and it has also been implicated as a stimulator of microglial mitosis after acute injuries (Raivich et al., 1991, 1994). From a functional point of view, it can be reasonably surmised that the rapid proliferation of microglial cells shortly after a CNS lesion occurs because greater numbers of these cells are required to initiate the complex processes of wound healing and tissue repair (Streit et al., 2000). Microglial activation is also thought to occur with norma l aging and in age-related neurodegenerative diseases, such as Alzh eimers disease (AD) (McGeer et al., 1987; Rogers et al., 1988; Streit and Sparks, 1997; Akiyama et al., 2000; McGeer and McGeer, 2001). However, unlike the microgliosis observe d after acute CNS inju ries, there is no evidence to show that ageor AD-rela ted microglial activation is accompanied by increased cell division. Quite to the contrary, there is evid ence showing that increased 85

PAGE 99

86 microglial cell death, as well as microglial structural abnorm alities (microglial dystrophy) are prominent features of the aged and AD brain (Lassmann et al., 1995; Yang et al., 1998; Jellinger and Stadelmann, 2000; Streit et al., 2004a). The latter observations have raised the possibility that a loss of microgl ial cells or of microglial cell function could contribute to the development of age-related neurodegenerati ve diseases (Streit, 2002a, b). Notwithstanding these relatively recent findings, the notion that chronic microglial activation (often referred to as neuroinflammation) is de trimental and a contributing factor in AD pathogenesis has become widely accepted and has resulted in the use of anti-inflammatory regimens as potential treatments (Akiyama et al., 2000). The neuroinflammation concept has fueled the idea that oxidative stress increases in the aging brain, in part because activated microglia in vitro produce reactive oxygen species (Colton and Gilbert, 1987), and it is therefore not surprising that antioxidants have been, and continue to be, explored as potential an ti-aging treatments (Jackson et al., 1988; Sano et al., 1997; O'Donnell and Lynch, 1998; Milg ram et al., 2002; Devi and Kiran, 2004). Function of Vitamin E Vitamin E is the most effective lipid-soluble antioxidant in biological membranes, and it acts to stabilize lipid membranes and prevent propagation of free radical damage (Halliwell and Gutteridge, 1985). Vitamin E (i.e., alpha-tocopherol) has an organic structure possessing two aromatic rings and a hydrocarbon tail. Cell surface receptors exist for the binding and uptake of vitamin E (Kaempf-Rotzoll et al., 2003; Meier et al., 2003), however, very few studies have examined which specific receptors are responsible for its uptake. Specific receptor sites have been found for vitamin E on bovine adrenocortical cells (Kitabchi et al., 1 980) human erythrocytes (Kitabchi and Wimalasena, 1982), and rat hepatocytes (Murphy and Mavis, 1981). Dietary

PAGE 100

87 supplementation of vitamin E has shown bene fits for immune cell function (Bendich, 1988; Tengerdy, 1989), as well as cognitive perf ormance and neuroprotection (Socci et al., 1995; Perrig et al., 1997; Behl and Holsboe r, 1998; Joseph et al ., 1998; Joseph et al., 1999; Martin et al., 1999; Behl and Moosmann, 2002; Grundman and Delaney, 2002; Martin et al., 2002). There are complete d and ongoing human clin ical trials using vitamin E as a potential treatment for AD (Sa no et al., 1997). The effects of vitamin E on cultured microglial cells have been studied, a nd most of this work, consistent with the idea that microglial activation is a harmful pr ocess, has been interpreted to show that vitamin E suppresses microglial activation (Heppner et al., 1998; Li et al., 2001; Egger et al., 2001, 2003; Gonzalez-Perez et al., 2002; Godbout et al., 2004; Grammas et al., 2004). Thus, there appears to be a consensus currently that vitamin E may provide some neuroprotection by deactivating microglial cells. In the current study, we have investigated long term effect s of vitamin E on primary rat microglial cell cultures with the goal of determining its effects on cellular ag ing and proliferation ki netics. The results reveal that vitamin E is a most potent microglial mitogen that stimulates dramatic microglial proliferation in vitro Not unexpectedly, we have also found concomitant shortening of telomere length and loss of telomerase activity in these cultures. Materials and Methods Culturing of Microglia Microglia were isolated as described in Chapter 1. Treatment of Microglial Cells Microglia were treated on day 0 with either 10.2 nM (0.15 g/mL) recombinant rat granulocyte-macrophage colony stimulati ng factor (GM-CSF) (R&D Systems, Minneapolis, MN), DL-tocopherol acetate (S igma-Aldrich, St. Louis, MO; catalog #

PAGE 101

88 T3376-25G) dissolved in 100% ethanol (EtOH) a nd sterile-filtered at concentrations of 20.0 M (9.5 g/ L), 105.7 M (50 g/ L), or 500.0 M (236 g/ L), DL-lipoic acid (Sigma-Aldrich, St. Louis, MO; catalog # T-1395) dissolved in 100% EtOH and sterilefiltered at concentrations of 20.0 M (9.5 g/ L), 105.7 M (50 g/ L), or 500.0 M (236 g/ L), -tocopherol and -lipoic acid (same concentrations as above), or an identical volume of sterile-filtered 10 0% EtOH as a vehicle control (i.e., 10 L/1 mL media). In all experiments, media (and respective treatment) was changed as needed (usually every 3 to 4 days). Determination of Cell Proliferation Cell proliferation was measured as described in Chapter 1. Determination of Interleukin-1 Production To determine production of interleukin-1 beta (IL-1 ) by cultured microglia, the Quantikine M rat IL-1 immunoassay was used (R & D Systems, Minneapolis, MN). Quantitative determination of rat IL-1 concentrations in cell culture supernatants was performed according to the manu facturers recommended protocol. Determination of Telomere Length Telomere length was measured as described in Chapter 1. Determination of Telomerase Activity Telomerase activity was measured as described in Chapter 1. Statistical Analysis of Data Statistical analyses were performed as described in Chapter 3.

PAGE 102

89 Results Microscopic Examination of Cultured Rat Microglia at Various Times and Treatments Representative photographs of cultured rat microglia on the indicated day under various treatment conditions are shown in Fig. 5-1. Microglia treated with 1% EtOH (vehicle controls) gradually decline in cell number from day 2 to 7 and were stable from day 7 to 14. The most dramatic effect is evident in microglia treated with 105.7 M tocopherol (vitamin E). These cells underwent rapid and sustained proliferation from day 0 to 2, compared to EtOH controls, peaked in cell number at around day 7, and then declined substantially in cell number fr om day 7 to 14. Cells treated with 500 M lipoic acid (LA) exhibited a gradual d ecline in cell number from day 2 to 14. Vitamin E Induces Cell Proliferati on in Cultured Rat Microglia Cell proliferation (as determ ined by MTT) of cultured rat microglia from day 2 to 27 under various treatment conditions is shown in Fig. 5-2. As reported before (Flanary and Streit, 2004), treatment of microglia with 10.2 nM recombinant rat granulocytemacrophage colony stimulating factor (GM-CSF) resulted in a signi ficant and sustained increase in cell proliferation compared to other treatments from day 2 to 27. Proliferation was significantly higher in GM-CSF-treated microglia (compared to all other groups, except vitamin E) on day 7 [p<0.001 (compared to EtOH vehicle control and LA-treated cells)], day 14 [p<0.05 (compared to vitamin E and E & LA-treated cells); p<0.001 (compared to EtOH vehicle control and LA-t reated cells)], and day 27 [p<0.01 (compared to EtOH vehicle control-trea ted cells); p<0.001 (compared to vitamin E, LA, and E & LA-treated cells)]. Treatme nt of microglia with 105.7 M -tocopherol (vitamin E)

PAGE 103

90 Figure 5-1. Representative mi crographs of cultured rat microglia under various treatment conditions: [100% EtOH (10 L/1 mL media) as vehicle control, 105.7 M -tocopherol acetate (i.e., vitamin E: Vit E), 500.0 M -lipoic acid (LA), or both Vit E and LA (same concentrations as above)]. A, B, C = 100% EtOHtreated cells on days 2, 7, and 14; D, E, F = Vit E-treated cells on days 2, 7, and 14; G, H, I = LA-treated cells on days 2, 7, and 14; J, K, L = E & LAtreated cells on days 2, 7, and 14. All photographs are at 10X magnification. Scale bar = 20 m. produced the greatest significant increase in microglial proliferation from day 2 to 7 (p<0.001), compared to all other treatment groups. However, on day 14, proliferation dropped to levels below that of GM-CSFtreated microglia, only to peak again significantly on day 21 [p<0.01 (compared to E & LA-treated cells); p<0.001 (compared

PAGE 104

91 to EtOH vehicle control and LA -treated cells), and subse quently decline on day 27. These results correlate well wi th the photographs presented in Fig. 5-1 in that microglia were shown to increase in cell number from da y 2 to 7 and decrease substantially in cell number from day 7 to 14. The combination treatment of vitamin E and -lipoic acid (E & LA) resulted in an intermediate cell proliferation between that of -lipoic acid (LA) or vitamin E alone. Microglia treated with LA exhibited cell proliferation below that of EtOH controls on all days examined. On day 2, microglial cells treated with E & LA had a significant higher cell proliferation than LA alone (p<0.05), while on day 14 exhibited a higher proliferation than both EtOH controls (p<0.001) and LA (p<0.001). Figure 5-2. Cell proliferation (as determined by MTT assay) of cultured rat microglia on the indicated days under various tr eatment conditions: [100% EtOH (10 L/1 mL media) as vehicle control, 105.7 M -tocopherol acetate (i.e., vitamin E: Vit E), 500.0 M -lipoic acid (LA), both Vit E and LA (E & LA) (same concentrations as above), or 10.2 nM recombinant rat granulocytemacrophage colony stimulating factor (GM-CSF)]. Treatment with 105.7 M vitamin E produced the greatest incr ease in microglial proliferation.

PAGE 105

92 The proliferation rate (as determined by BrdU incorporation) of cultured rat microglia from day 1 to 12 under various treatment conditions is shown in Fig. 5-3. Treatment of microglia with 10.2 nM recombinant rat granulocyte-macrophage colony stimulating factor (GM-CSF) re sulted in a sustained significant increase in proliferation rate from day 2 to 12 compared to other tr eatments (except vitamin E-treated microglia from day 2 to 12). Treatm ent of microglia with 105.7 M -tocopherol (vitamin E) produced the greatest in crease in proliferation rate from day 2 to 12. Proliferation rate was significantly higher in vitamin E-treated microglia (compared to all other groups) on day 5 [p<0.01 (compared to EtOH vehicle c ontrol and GM-CSF-treated cells); (p<0.001 compared to LA and E & LA-treated cells)], day 7 (p<0.001 comp ared to all other treatments), and day 12 (p<0.001 compared to all other treatment s). These results correlate well with the data presented in Fi g. 5-1 and 5-2. The combination treatment of E & LA resulted in an intermediate prolifera tion rate between that of LA or vitamin E alone. Microglia treated with LA exhibited proliferati on rates below that of EtOH controls on all days examined, as was seen in the MTT assay. Quantitation of cell proliferation of culture d rat microglia at 48 hours under various treatment conditions is shown in Fig. 54. Treatment of microglia with 20 M vitamin E resulted in the highest increas e in proliferation compared to all other treatment groups and concentrations. Notwithstanding 20 M E & LA, treatment of microglia with 20, 105.7, and 500 M vitamin E yielded the top 3 highest increases in ce ll proliferation, respectively, compared to other treatment groups. Prolifer ation was significantly higher (p<0.01 to 0.001) in 20 M vitamin E-treated microglia, compared to all other groups (except 20 M E & LA-treated cells). Tr eatment of microglia with 20 M E and LA

PAGE 106

93 Figure 5-3. Proliferation rate (as determined by BrdU in corporation over 2 hours) of cultured rat microglia on the indi cated days under various treatment conditions: [100% EtOH (10 L/1 mL media) as vehicle control, 105.7 M -tocopherol acetate (i.e., vitamin E: Vit E), 500.0 M -lipoic acid (LA), both Vit E and LA (E & LA) (same c oncentrations as above), or 10.2 nM recombinant rat granulocyte-macropha ge colony stimulating factor (GMCSF)]. Treatment with 105.7 M vitamin E produced the greatest increase in microglial proliferation, and was signif icantly higher than EtOH, LA, E&LA, and GM-CSF on days 5, 7, and 12 (p<0.001). resulted in the second-highest increase in proliferation in all instances, and was significantly higher (p <0.01 to 0.001) than all other treatment groups (except 20 M and 105.7 M vitamin E-treated cells). The combination treatment of E & LA resulted in an intermediate proliferation between that of LA or vitamin E alone. In all treatment groups, there was a strong inverse corr elation between proliferation and treatment concentration, such that cell proliferation wa s highest when treated with th e lowest concentration (i.e., 20 M) of each treatment, and vice versa.

PAGE 107

94 Telomere Length Analysis in Vitamin E-Treated Cultured Rat Microglia Quantitation of telomere length (via Southe rn blot analysis) in microglia under various treatment conditions from day 0 to 7 is shown in Fig. 5-5. In EtOH controls, both the longest and shortest telomeres underwent little change from day 0 to 7, while the mean telomeres shortened from day 0 to 2, a nd subsequently lengthened from day 2 to 7. In both vitamin E-treated and -lipoic (LA) acid-treated microglia, the longest telomeres also underwent little change from day 0 to 7, with the shortest telomeres gradually decreasing in length, while the mean telomeres underwent a moderate/l arge (respectively) Figure 5-4. Proliferation of cultured rat microglia at 48 hours under various treatment conditions: [100% EtOH (10 L/1 mL media) as vehicle control, 20.0, 105.7, or 500.0 M -tocopherol acetate (i.e., vitamin E: Vit E), 20.0, 105.7, or 500.0 M -lipoic acid (LA), or both Vit E and LA (E & LA) (same concentrations as above)]. Treatment with 20 M vitamin E produced the greatest increase in mi croglial proliferation. decrease in length from day 0 to 7. On da y 7 in microglia treated with LA, the mean telomeres were over 100% shorter compared to EtOH controls on the same day. In microglia treated with a combination of vitamin E and LA, the shortest telomeres

PAGE 108

95 remained relatively unchanged, however, both th e largest and mean telomeres exhibited large decreases in length from day 2 to 7 and day 0 to 7, respectively. On day 7 in microglia treated with vitamin E and LA, bot h the longest and mean telomeres were well over 100% shorter compared to EtOH controls on the same day. Overall, telomere lengths were shorter on both day 2 and 7 in mi croglia in all treatment groups compared to EtOH controls on corresponding days. 0 5 10 15 20 25 30 35 40Telomere Length (kb) Longest Mean Shortest Longest 30.3231.9133.8431.7432.5331.9730.1330.3214.17 Mean 24.3813.7718.7423.9714.7021.839.037.738.74 Shortest 5.154.93 4 .36 4 .69 4 .225.20 4 .694.575.28 Day 0 Day 2 EtOH Day 7 EtOH Day 2 Vit. E Day 7 Vit. E Day 2 LA Day 7 LA Day 2 E&LA Day 7 E&LA Figure 5-5. Densitometric quantitation of telome re length in cultured rat microglia on the indicated days under various treat ment conditions: [100% EtOH (10 L/1 mL media) as vehicle control, 105.7 M -tocopherol acetate (i.e ., vitamin E: Vit E), 500.0 M -lipoic acid (LA), or both Vit E and LA (same concentrations as above)]. Telomerase Activity Analysis in Vitamin E-Treated Cultured Rat Microglia A representative photograph of a TRAP analysis image used for determination of telomerase activity in cultured rat microgl ia is shown in Fig. 5-6. Quantitation of telomerase activity in microglia under various treatment conditions from day 0 to 14 is shown in Fig. 5-7. In EtOH controls, telo merase activity significantly increased

PAGE 109

96 (p<0.001) from day 0 to 7, then significantly declined (p<0.001) from day 7 to 14. Treatment with -lipoic acid (LA) also caused an in itial significant increase (p<0.001) in telomerase activity from day 0 to 7, with a subsequent signifi cant decrease (p<0.001) from day 7 to 14. The pattern of telomerase activity in LA-treated cells mirrored that in EtOH controls, and LA-treated ce lls exhibited lower levels of activity at all times points relative to controls. In vitamin E-treated microglia, telomerase activity significantly increased (p<0.001) from day 0 to 2, declined from day 2 to 7, then increased again from day 7 to 14. In microglia treated with a co mbination of vitamin E and LA, telomerase activity significantly declined (p<0.001) from day 0 to 14. In relation to EtOH controls, every treatment suppressed telomerase activ ity except for vitamin E on day 14. The only treatment that yielded an overall increase in telomerase activity from day 0 to 14, and was the treatment that resulted in the highest te lomerase activity at day 14, was vitamin E. Quantitation of telomerase activity in cultured rat microglia at 48 hours under various treatment conditions is shown in Fi g. 5-8. Telomerase activity was high in vitamin E-treated microglia at 20 and 105.7 M concentrations, but significantly decreased (p<0.001) by over twofold when treated with 500 M vitamin E. Telomerase activity was also high in mi croglia treated with 105.7 M -lipoic acid (LA), but was approximately 33% and 50% lower (p<0.001) in cells treated with 20 uM and 500uM LA, respectively. In microglia treated w ith a combination of vitamin E and LA, telomerase activity was moderately high at 20 M concentrations, but significantly decreased (p<0.001) by approximately tw o-fold when treated with 105.7 or 500 M LA. In all instances, telomerase activity decreased with increa sing treatment concentration, except in LA-treated microglia from 20 to 105.7 M.

PAGE 110

97 Figure 5-6. Representative TR AP analysis image used for measurement of telomerase activity in cultured rat microglia. Shown here is the TRAP image of microglia at 48 hours under various treatment conditions: [100% EtOH (10 L/1 mL media) as vehicle control, 20.0, 105.7, or 500.0 M -tocopherol acetate (i.e., vitamin E: Vit E), 20.0, 105.7, or 500.0 M -lipoic acid (LA), or both Vit E and LA (same concentrations as above)] Neg = telomerase-negative control; Pos = telomerase-positive cont rol; IC = internal control.

PAGE 111

98 Figure 5-7. Quantitation of telomerase activity in cultured microglia on the indicated days under various treatment conditions : [100% EtOH (vehicle control), 105.7 M -tocopherol acetate (i.e., vitamin E: Vit E), 500.0 M -lipoic acid (LA), or both Vit E and LA (sam e concentrations as above)]. Interleukin-1 Beta Production in Cultured Rat Microglia Quantitation of IL-1 production by cultured microgl ia under various treatment conditions from day 2 to 7 is shown in Fig. 59. Treatment of microglia with either EtOH (control), vitamin E, -lipoic acid (LA), or E & LA caused a significant increase (p<0.001) in production of IL-1 from day 2 to 7. In microglia treated with a combination of vitamin E and LA, production of IL-1 was significantly higher (p<0.001) compared to all other treatments at each time point measured. In addition, each treatment group was significantly higher (p<0.001) than EtOH controls on each day, except for vitamin E alone and LA alone on day 7. Importantly, IL-1 production in vitamin E-treated microglia was significan tly lower than all other groups on day 7.

PAGE 112

99 Figure 5-8. Quantitation of telomerase activ ity in cultured microglia at 48 hours under various treatment conditions: [100% EtOH (vehicle control), 20.0, 105.7, or 500.0 M -tocopherol acetate (i.e., vitami n E: Vit E), 20.0, 105.7, or 500.0 M -lipoic acid (LA), or both Vit E a nd LA (same concentrations as above)]. Figure 5-9: Quantitation of interleukin-1 production by cultured rat microglia on the indicated days under various treat ment conditions: [100% EtOH (10 L/1 mL media) as vehicle control, 105.7 M -tocopherol acetate (i.e ., vitamin E: Vit E), 500.0 M -lipoic acid (LA), or both Vit E and LA (same concentrations as above)]. A rat IL-1 positive control (included in the kit) yielded 174.5 pg/mL.

PAGE 113

100 Discussion This study is the first to report the mitogenic effects of -tocopherol on microglia, and it shows that treatment of cultured rat micr oglia with vitamin E for 7 days resulted in a very significant increase in cell numbers. The magnitude of the mitogenic effect of vitamin E on microglia is approximately tw ice that of the known microglial mitogen, GM-CSF, and thus, vitamin E may be consid ered to be at present the most potent microglial mitogen known. As expected, and in line with our prior work (Flanary and Streit, 2004), the increase in microglial pro liferation was accompanied by a decrease in telomere length in the presen ce of decreasing telome rase activity. In addition, our results show that the mitogenic effect of vitamin E on microglial proliferati on is independent of its anti-oxidative action since cultures trea ted in parallel with the antioxidant -lipoic acid did not show a proliferative response. Microglia treated with vitamin E underwent a massive increase in cell number from day 2 to 7 and then declined from day 7 to 14, correlating well with BrdU incorporation over the same time period. These findings unders core the transient natu re of this effect in vitro and beg the question whether vitami n E can sustain microglial mitosis in vivo in the long term. In vitro, vitamin E-induced microglial proliferation falls to levels below those of controls on day 27, supporting the notion that these cells have exhausted their replicative capacity at this point, and are like ly nearing replicative senescence. A similar situation existed for microglia treated with vitamin E & LA in that they, too, exhibited proliferation that was significantly higher than in EtOH controls, however, following day 7 this proliferation declined steadily and eventu ally fell below levels seen in controls on day 27. In LA-treated microglia, cell proliferation was lower than in controls at all time

PAGE 114

101 points examined. In addition, cell proliferation in E & LA-t reated microglia was lower compared to vitamin E-treated cells on all days examined, and was significantly lower (p<0.001) than vitamin E-treated cells on day 2, 7, and 21. Collectively, these observations suggest that LA suppressed mi croglial proliferation. On the other hand, GM-CSF-treated microglia also exhibited an early significant increase in cell proliferation compared to vehicle controls, yet these cells were able to sustain their proliferative activity through day 27. This suggests that GM-CSF-treated microglia may be able to proliferate longer before enteri ng replicative senescence compared to vitamin E-treated microglia, perhaps because GM-CSF-tr eated microglia divided at a lower level relative to vitamin E-treated cells. Vitami n E-treated microglia seemingly used up most of their proliferative poten tial during the first week of in vitro growth. In our previous studies (Flanary and Streit, 2004), we found that GM-CSF-treated cultured rat microglia exhibite d higher cell proliferation, gr eater telomerase activity, and increased telomere length compared to controls during the first 2 days of in vitro growth, presumably in an effort to compensate for the loss of telomeric repeats that would otherwise occur during the period of high pr oliferation. Vitamin E-treated microglia exhibited an even greater incr ease in proliferatio n from day 2 to 7 than GM-CSF-treated cells, and also showed an increase in telome rase activity. However, telomere length in vitamin E-treated microglia decreased from da y 0 to 7, suggesting that the very high rate of proliferation caused by vitamin E exposur e resulted in the inexorable shortening of telomeres despite an increas e in telomerase activity dur ing the same time period. The processivity of telomerase (Greider 1991), as determined by banding pattern on gel, correlated well with overall telome rase activity. Lanes possessing an intense

PAGE 115

102 initial band (immediately above the internal control band), which always contained a large quantity of small telomerase products and a small quantity of larger products, represent samples with low enzyme processivity (e.g., 500 M vitamin E). The presence of an intense initial band indicated that most of the time telomerase added only a single hexanucleotide telomeric repeat onto the end of the TS primer before disassociating. Thus, a great proportion of the overall measur ed telomerase activity was contributed by the presence of the first few initial bands. In these lanes, telomerase could only generate small products and lacked the ability to create larger ones (which are clearly present in lanes with high telomerase processivity). The presence of low enzyme processivity usually coincided with lowe r overall telomerase activity in our samples. Production of interleukin-1 one type of cytokine that mediates the inflammatory response, was significantly decreased in vitamin E-treated microglia in vitro. This suggests that vitamin E significantly suppressed microglial activation in vitro in terms of IL-1 production (on day 7), and is consistent with previ ous research showing that vitamin E may provide neuroprotection by s uppressing intracellula r signaling events necessary to induce microglial act ivation (Heppner et al., 1998; Li et al., 2001; Egger et al., 2001, 2003; Gonzalez-Perez et al., 2002; Godbout et al., 2004; Grammas et al., 2004), and that vitamin E decreases/suppresses IL -1B production (Akeson et al., 1991; Devaraj et al., 1996; Devaraj and Jialal, 1999; Pathan ia et al., 1999; Gon zalez et al., 2001). Interleukin-1 production was significantly higher in E & LA-treated microglia compared to all other treatments at each timepoint measured, despite the significantly lower IL-1 levels produced with either vitamin E or LA treatments alone. This suggests that a synergistic effect occurr ed (in terms of increased IL-1 production) when both

PAGE 116

103 antioxidants were used in combination (S cholich et al., 1989; Haramaki et al., 1993; Stoyanovsky et al., 1995), that a combinational treatment is more efficient at activating microglia in vitro and that LA decreased/eliminated the ability of vitamin E alone to significantly lower IL-1 levels. In all instances, IL-1 production increased from day 2 to 7, suggesting that IL-1 levels increase as cells age in vitro and progress toward senescence. This correlates with previous re search indicating that activated interleukin-1 positive human microglia increase with age (Sheng et al., 1998. A wealth of previously-pub lished research supports our current findings (i.e., that vitamin E exerts mitogenic effects). One of the earliest reports s howed that vitamin E enhanced guinea pig smooth muscle cell pro liferation (Miller et al., 1980). Previous research has also indicated that vitamin E acts as a potent mitogen for peripheral macrophages (Oonishi et al., 1995) and lymphocyt es (Meydani et al., 1986; Roy et al., 1991; Sakai and Moriguchi, 1997). Clearly, vitamin E plays an important role as a mitogenic agent in a variety of cell types. Surprisingly, none of the studies published thus far have reported the in crease in microglial prolifera tion that we observed. Heppner et al. (1998) noted that cell numbers of microglia treated w ith vitamin E remained stable within 7 days in vitro and that vitamin E was shown to protect microglial cell cultures from excessive loss of viable cells ; however, this study only used 10 g/mL vitamin E treatment, which may account for the absence of microglial prolifer ation. Heppner et al. (1998) concluded that since vi tamin E treatment of primary rat microglia cells caused them to down-regulate expression of adhesion molecules associated with microglial activation, that scavenging of free radicals (via vitamin E) may prevent/reverse microglial activation. Similarly, a study by Li et al. (2001) found that vi tamin E treatment inhibited

PAGE 117

104 LPS-induced activation in the N9 murine microglial cell line as determined by downregulation of inflammatory cytokines (i.e., IL-1 TNF), and nitric oxide via a p38 mitogen-activated protein kinase and nucle ar factor kappa B-dependent pathway; however there is no mention of changes in microglial cell numbers following vitamin E treatment. Studies by Egger et al. (2001, 2003) found that treatment of both primary porcine microglia and the BV-2 murine micr oglial cell line with vitamin E (up to 100 M) attenuated superoxide production and tr anscription of cyclooxygenase 2; however, there is no mention in either report of changes in microglial cell numbers following vitamin E treatment. Other studies have found that vitamin E treatment reduced LPSinduced activation in primary murine micr oglial cells as determined by decreased production of the inflammatory cytokine IL -6 (Godbout et al., 2004) and prostaglandin E(2) production (Grammas et al., 2004), but make no reference to potential mitogenic effects. In the current study, the mitogenic effect of vitamin E on microglia is most dramatic in that it exceeds the effect of the well-known microglial mitogen, GM-CSF (Giulian and Ingeman, 1988) and thus, vitamin E may be considered to be (at present) the most powerful known mitogen for microglia in vitro. We have been able to consistently and repeatedly reproduce this effect using different batches of both microglial cultures, and alpha tocopherol lots/stock solutions, in our laboratory. Even though the CNS is generally considered a post-mitotic tissue, it is important to note that microglia do retain a robust proliferative potential, especially under conditions of CNS injury, as shown by studies using 3 H-thymidine incorporation (Graeber et al., 1988; Kreutzberg, 1996). This increase in microglial cell numbers followi ng neuronal injury likely occurs because

PAGE 118

105 additional cells are needed to assist injured neurons (Str eit, 2002b). Vitamin E may therefore act as a neuroprote ctive agent by not only suppre ssing microglial activation (in terms of cytokine production), but also by increasing microglial cell numbers. Although an increase in cell numbers is one aspect of microglial activation, this feature by itself may not be a harmful or detrimental proce ss to microglia themselves or to bystander cells. If vitamin E does result in an increase in microglial proliferation in vivo then this may provide an additional mechanism (in addi tion to scavenging free radicals) to explain why vitamin E supplementation has been shown to be beneficial in rodent models of stroke (Stohrer et al., 1998; van der Worp et al., 1998; Ta gami et al., 1999; Noguchi et al., 2001; Gonzalez-Perez et al., 2002; Chaudhary et al., 2003; Ikeda et al., 2003; Niu et al., 2003) and brain injury (Yoshida et al., 1985; Clifton et al., 1989; Inci et al., 1998; Ikeda et al., 2000; Conte et al., 2004). In the current study, vitamin E was shown to induce microglial proliferation with concomitant telomere shortening in vitro. However, we have found that telomeres lengthen in axotomy-activated proliferating rat microglia in vivo (Flanary and Streit, submitted). Although it remains unknown what effects, if any, vitamin E has on microglia in vivo it remains possible that vitamin E supplementation in rodents/humans may also induce a similar high pr oliferative rate of microglia in vivo If proliferation occurs, this may result in an accelerated rate of telomere shortening in microglia, in particular because all examined human somatic cell types do not express telomerase activity (Kim et al., 1994), and thus may hasten their entry into replicative senescence (Flanary, 2004; Fossel, 2004). However, telo merase activity has been detected in human lymphocytes and hematopoietic progenitor cells (Hiyama et al., 1995). Since human

PAGE 119

106 microglia originate from bone marrow-derive d hematopoietic progenitor cells (Eglitis and Mezey, 1997; Hess et al., 2004), they too may be able to upregulate telomerase activity in certain instances (and thereby possibly prevent/slow telomere shortening following periods of rapid division). The fi ndings from this study provide an impetus to further investigate telomere biology in hu man microglia, as well as the roles of in vivo vitamin E supplementation in rats with age, and following CNS injury, in relation to telomere dynamics and microglial cell proliferation.

PAGE 120

CHAPTER 6 LIFE-SPAN EXTENSION IN NORMAL RAT MICROGLIA VIA TELOMERASE REVERSE TRANSCRIPTASE RETROVIRAL TRANSDUCTION Introduction Telomeres in rat microglia (the only speci es examined to date) have been found to shorten with age when cultured in vitro (Flanary and Streit, 2 004). Thus, in order to determine whether exogenous delivery of the telomerase gene via re troviral transduction could prevent microglial senescence and exte nd the life-span of cultured rat microglia, we decided to measure parameters indicative of telomeric maintenance, such as telomere length and telomerase activity. Materials and Methods Culturing of Microglia Microglia were isolated as described in Chapter 1. Production of Replication-Defectiv e Telomerase-Encoding Retroviruses The retroviral vector, pLPC-hTRT, was obtai ned via a material transfer agreement (reference #: 3354/MTA hTRT/Streit) through Geron Corporation (Menlo Park, CA) and Clontech (Palo Alto, CA). The pLPC-h TRT retroviral vector expresses human telomerase reverse transcriptase (hTRT) fr om the cytomegalovirus promoter. The 5 viral long terminal repeat controls expr ession of the transcript that contains + (the extended viral packaging signal), the puromycin resistance gene for antibiotic selection in eukaryotic cells, and telomerase. Vector DNA was amplified in DH5 competent cells (Invitrogen, Carlsbad, CA, catalog #: 18263-012) in the presence of ampicillin 107

PAGE 121

108 (resistance was conferred by the presence of pLPC-hTRT, which contained an ampicillinresistance gene), purified using the hi-speed maxi prep kit (Qiagen, Valencia, CA, catalog #: 12662), and stored at -20 C until needed. DNA concentration was determined by absorbance at 260 nm, while DNA purity was ca lculated by the ratio of 260 nm versus 280 nm absorbance, using a spectrophotometer. The Phoenix amphotropic retroviral packaging cell line (a gift from Michel Ouelle tte, Ph.D., University of Nebraska Medical Center, Lincoln, NE) was used to produce telo merase-encoding retrovirus. Phoenix cells were thawed from cryovials stored in liqui d nitrogen, expanded in DMEM, passaged via trypsinization, plated into 21 cm 2 dishes (2.5 to 5.0 x 10 5 cells per dish), and allowed to divide overnight at 37 C under 5% CO 2 The following morning, cells that were 80 to 90% confluent were transfected with 10 to 20 g of either: 1) pLPC-hTRT plasmid DNA, 2) empty vector (i.e., pLPC vector on ly), or 3) green fluorescence protein (GFP) vector (Clontech vector pLEG FP-C1, a gift from John Rossi at the Beckman Research Institute of the City of Hope, Duarte, CA) using the CaPO 4 -mediated MBS mammalian transfection kit (Stratagene, La Jolla, CA ) according to the manufacturers recommended transfection protocol. Retr ovirus was harvested over th e next 24 to 72 hours posttransfection by collec ting the supernatant, and fi lter-sterilizin g it using a 0.45 m PVDF filter (Fisher Scientif ic, Pittsburgh, PA, catalog #: 09-7204). Viral supernatants were either stored on ice and used immediat ely, or aliquoted and stored at -80 C (for up to 6 months) until used. Permission to use both the pLPC-hTRT retrovir al vector and the replication-defective telomerase-encoding re troviruses was approved by the University of Florida Environmental Health and Safety offi ce (project approval number: RD-2436).

PAGE 122

109 Transduction of Rat Microglia With Telomerase-Encoding Retroviruses Rat microglia, as well as a rat glioblastoma cell line (i.e., RG-2) were plated (on day 0) in 9.5 cm 2 plates (1.5 to 2.5 x 10 5 cells/well) and allowed to divide overnight at 37 C under 5% CO 2 The following morning, media was aspirated, and retroviral transduction (using freshly-harvest retrovi rus) was accomplished (on day 1 or 2) by adding a mixture containing 1 volume of vira l supernatant, 1 volume of DMEM, and 4 g/mL polybrene (Sigma-Aldrich, St. Louis, MO, catalog #: H-9268) directly to the cells. Wells containing microglia also had 10.2 M (0.15 g/mL) of the mitogen recombinant rat granulocyte-macrophage colony stimulating factor (CSF) (R&D Systems, Minneapolis, MN, catalog #: 518-GM ) added to the media to help stimulate cell proliferation (as required by retroviruse s for transduction to occur). Infections performed included: 1) cells infected with pLPC-hTRT, 2) cells infected with pLPC empty vector, 3) cells infected with GFP vect or (pLEGFP-C1A) or 4) cells infected with virus-free media (i.e., non-transduced). The pLPC-hTRT vector transiently expresses, or integrates and stably expresses, a transc ript containing the ex tended viral packaging signal ( + ), the puromycin resistance gene (Puro R ), and human telomerase reverse transcriptase (hTRT). Only the protein co mponent of telomerase was needed during transduction, since when the protein co mponent binds with normal endogenous RNA subunits of telomerase, a functional enzyme will result. Transduction efficiency was monitored by transducing both rat microglia and RG-2 cells with a GFP vector (pLEGFPC1, Clontech, Palo Alto, CA), and analyzi ng the cells for subsequent GFP expression using a Zeiss Axiovert 25 fluorescence inverted microscope. Within 24 to 72 hours, cells were infected sequentially with each of the different harvests of the same virus and

PAGE 123

110 allowed to divide overnight (16 hours) at 37 C under 5% CO 2 Following transduction, CSF was added to microglia to stimulate cell pr oliferation and aid in the incorporation of the viral genome. Cells were selected for stable viral integrations using puromycin (Sigma-Aldrich, St. Louis, MO, catalog #: P8833) at concentrati ons of 500 ng/mL for microglia. Puromycin was added to the media of all transduced cells (except for one set of cells that was not transduced at all) for a pproximately 7 days, or until all cells infected with virus-free media were dead. Transdu ced microglia were cultured in either the continual presence or absence of CSF. Ce lls were photographed at various time points using a digital camera (Sony DSC-S75 Cyber-s hot, 3.3 megapixels, Carl Zeiss VarioSonnar lens) connected to a Zeiss Axiovert 25 fluorescence inverted microscope. Determination of Telomerase Activity Telomerase activity was measured as described in Chapter 1. Telomerase Western Blot Analysis Telomerase protein quantity was measured as described in Chapter 3. Statistical Analysis of Data Statistical analyses were performed as described in Chapter 3. Results Telomerase-Encoding Retroviral Vector The map of the retroviral v ector used in transfection and transduction experiments is shown in Fig. 6-1. The pLPC-hTRT retrovira l vector contains elements derived from the Moloney murine leukemia virus, a nd expresses human telomerase reverse transcriptase (hTRT) driven from the cytomega lovirus promoter (CMV IE). The 5 viral long terminal repeat (LTR) is the viral promoter that controls expres sion of the transcript containing + the viral packaging signal (required for encapsidation of the vector), the

PAGE 124

111 puromycin resistance gene (Puro r ), which is used when selecting mammalian cells posttransduction, the telomerase gene (hTRT), a nd the 3 viral LTR, which encodes the polyA signal. The ampicillin-resistance gene (Amp r ), is used for propagation and selection in bacteria, and the Col E1 is the origin of replication during bacterial propagation. Telomerase Transduction Exte nds Life-Span of Microglia Transduction efficiency was monitored by transducing both rat microglia and a rat glioblastoma cell line (i.e., RG-2) with a GFP vector (pLEGFP-C1), and analyzing the cells for subsequent GFP expression (Fig. 62). In both GFP-transduced microglia and RG-2 cells, there was evidence that trans duction was successful, since both cell types exhibited numerous green fluorescent cel ls in the days immediately following transduction. The efficiency of transducti on was estimated to be approximately 10%. Microglia transduced with te lomerase lived longer than both controls and pLPC (nontransduced) and empty-vector transduced microglia were all dead by day 20 (transduction day = day 1) (Fig. 6-3, 6-4). However, mi croglia transduced with the telomeraseencoding retrovirus were able to live until da y 75, representing an increase in maximal life-span of 375%. In a follow-up experiment in which microglia were cultured in the presence of CSF only during the first few da ys following transduction, both controls and empty-vector transduced microglia were all dead by day 27. However, microglia transduced with telomerase survived until day 62, representing an increase in maximal life-span of 230%. In both experiments, telomerase-transduced microglia gradually slowed in their cell division rate, an d eventually stopped dividing around day 50.

PAGE 125

112 Figure 6-1. The Clontech retrov iral vector, pLPC-hTRT, used to transduce cultured rat microglia with the human telomerase re verse transcriptase (i.e., hTRT) gene. The 5 viral long terminal repeat (LTR) contains promoter/enhancer sequences that control ex pression of the puromycin resistance gene (Puro R ) for antibiotic selection in eukaryotic cel ls. Expression of hTRT is driven by the human cytomegalovirus (CMV) immediate early promoter (P CMV IE ).

PAGE 126

113 Figure 6-2. Brightfiel d and green fluorescence microgra phs of rat microglia and rat glioblastoma cells (RG-2) following retroviral transduction on day 4. A: rat microglia brightfield image (arrow indicates GFP-positive cell); B: rat microglia fluorescence image (of A); C: RG-2 brightfield image (arrows indicate GFP-positive cells); D: RG-2 fluorescence image (of C). Note the GFP expressing microglia and RG-2 cells, indicating that transfection and subsequent transduction was successful. Scale bar = 20.0 m. Telomerase Activity in Transduced Microglia TRAP analysis was used for determination of telomerase activity in transduced rat microglia (Fig. 6-5). Telomerase activity increased from day 0 to 2, and declined significantly (p<0.001) from day 2 to 6 in c ontrols and hTRT-transduced microglia (Fig. 6-6). From day 6 onward, telomerase activity remained relatively unchanged in these two groups. In pLPC (empty vector)-transduc ed microglia, telomerase activity also

PAGE 127

114 Figure 6-3. Representative mi crographs of cultured rat microglia following retroviral transduction on days 1 and 20. A to C: Day 1; D to F: Day 20. A and D = non-transduced microglia (control); B a nd E = pLPC empty vector-transduced microglia; C and F = hTRT-transduced microglia. By day 20, both control and empty vector-transduced cells were all dead, whereas microglia transduced with hTRT were still al ive on the same day. Scale bar = 20.0 m.

PAGE 128

115 Figure 6-4. Representative photographs of hTRT-transduced cultured rat microglia on days 57 and 75. A to C: Day 57; D to F: Day 75. By day 75, only a very small percentage of microglia transduced with hTRT were still alive. Scale bar = 20.0 m.

PAGE 129

116 declined significantly (p<0.001) from day 2 to 3, increased moderately from day 3 to 6, then declined significantly (p<0.001) from day 6 to 22. In this experiment, both controls and empty-vector transduced microglia were all dead by day 27, whereas hTRTtransduced microglia survived for 62 days and exhibited telomerase activity at least up to day 37. Figure 6-5. TRAP analysis image used fo r measurement of telomerase activity of cultured rat microglia on the indicated days under various transduction conditions. Microglia were harvested on day 0, and transduced on days 2 and 4. Both controls and empt y-vector transduced microgl ia were all dead by day 27, whereas hTRT-transduced microglia survived until day 62. Con = nontransduced; pLPC = empty vector (p LPC)-transduced, hTRT = telomerasetransduced; Neg = telomerase-negative control; Pos = telomerase-positive control; IC = internal control.

PAGE 130

117 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 02468101214161820222426283032343638 DayAverage Telomerase Activity Con pLPC hTRT Figure 6-6. Quantitation of telomerase activity in cultured rat microglia on the indicated days under various transduction conditions. Con = non-transduced; pLPC = empty vector (pLPC)-transduced, hTRT = telomerase-transduced. Telomerase Protein Quantity in Telomerase-Transduced Microglia Western blotting was used for determina tion of telomerase protein quantity in transduced rat microglia (Fig. 6-7). Quantita tion of telomerase protein is shown in Fig. 6-8. Telomerase protein increased from day 0 to 2, and gradually declined from day 2 to 22 in controls and empty vector-transduced microglia. In hTRT-transduced microglia, telomerase protein quantity declined significan tly (p<0.001) from day 2 to 6, then slightly increased from day 6 to 37. In this expe riment, both controls and empty-vector transduced microglia were all dead by day 27, whereas hTRT-transduced microglia survived for 62 days and s till produced telomerase prot ein at least up to day 37.

PAGE 131

118 Figure 6-7. Western blot image used for meas urement of telomerase protein in cultured rat microglia on the indicated days u nder various transduction conditions. Microglia were harvested on day 0, and transduced on days 2 and 4. Both controls and empty-vector transduced microglia were all dead by day 27, whereas hTRT-transduced microglia survived until day 62. Con = nontransduced; pLPC = empty vector (p LPC)-transduced, hTRT = telomerasetransduced. 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 02468101214161820222426283032343638 DayAverage Telomerase Protein Con pLPC hTRT Figure 6-8. Quantitation of telomerase protein quantity in cultured rat microglia on the indicated days under various transducti on conditions. Con = non-transduced; pLPC = empty vector (pLPC)-transduced, hTRT = telomerase-transduced.

PAGE 132

119 Discussion We have shown, for the first time, that exogenous delivery of the telomerase gene in normal rat microglia can extend their repl icative and maximal lif e-span. In addition, telomerase-tranduced microglia exhibited a delay in their entry to senescence, and maintained a normal phenotype following tr ansduction. Our findings suggest that exogenous delivery of the telomerase gene can extend maximal life-span in cultured rat microglia. Despite low efficiency, retroviral tran sduction was successful, as evidenced by GFP expression in both microglia and RG-2 ce lls. Transduction of rat microglia with catalytic component of telomerase (i.e., hTRT ) resulted in life-span extension compared to controls (non-transduced ) and empty-vector transduced cells. Telomerase activity/protein, albeit at low levels, were present in telomerase-transduced microglia well past the timepoint when control cells had all died. In ad dition, the pattern of telomerase activity and telomerase protein quantity corresponded well, suggesting that there is a synergistic effect between the two (as e xpected). There is an approximate 7fold decrease in protein quantity (as determin ed by Western blot analysis) in telomerasetransduced microglia compared to controls. Importantly, a similar level of telomerase activity exists in all groups, suggesting that the low le vels of telomerase protein present within transduced microglia are very active and have considerable catalytic activity. Telomerase transduced microglia cultured in the continual presence of recombinant rat granulocyte macrophage-colony stimulating factor (CSF) lived 375% longer than controls. Telomerase transduced microglia cu ltured in the absence of CSF (except during the transduction process) lived 230% longer than controls. Previous research from our laboratory has shown that CSF-treated rat microglia exhibit an increase in both cell

PAGE 133

120 proliferation and telomerase activity (Flanary and Streit, 2004). The transduced microglia cultured continuously with CSF may have lived longer (by about 20%) than those cultured in the absence of CSF due to the continual presence of the mitogen CSF, which may have acted to induce cell prolifer ation and thereby promote the retention of the transduced telomerase plasmid by the cells. In the absence of such mitogenic stimuli, the transduced cells may have lost the telo merase plasmid (or excised it from the genome if previously incorporated), since proliferation (and therefor e an increase in telomerase activity) was not required. In terestingly, control cells liv ed only until day 20 when cultured in continual presence of CSF. However, when they were cultured in CSF for only the first few days, the control cells lived until day 27, representing a 35% increase in maximal life-span (compared to those that lived until day 20). When the control cells were cultured in the continual presence of CSF, it likely induced an increase in cell proliferation (i.e., at a higher rate compared when CSF was limiting), and hence telomere shortening, which may have hastened their en try into replicative senescence. Thus, control cells cultured in the presence of CSF for the first few days only were not induced to divide as much, and as a result, had a slower rate of telomere attrition which may have delayed their entry into replic ative senescence. In addition, a parallel experiment (data not shown) was also conducted for rat astrocyt es telomerase transduction. Both controls and empty-vector transduced astrocytes were all dead by day 44, however, astrocytes transduced with telomerase lived until day 134 (305% increase in ma ximal life-span). Previous studies have shown that hTRT tr ansduction can be used to significantly extend the replicative life-span of human ce lls, resulting in immort alization (Bodnar et al., 1998). In the present study, hTRT tran sduction of rat microglia did not result

PAGE 134

121 immortalization, but rather only extended the life-span of th ese cells. Rat somatic cells normally express telomerase (Golubovskaya et al., 1997), and telomera se transduction to any rat cell type has never been reported previously, as near ly all previous experiments that delivered exogenous telomerase did so to human cell types. Similar to our results, previous studies in human myocytes (Di Donna et al., 2003) and human brain endothelial cells (HBECs) (Gu et al., 2003) have also found that telomerase tr ansduction only results in life-span extension, but not immortalization. Interestingl y, the maximum extension of life-span observed in human myocytes in the study by Di Donna et al. (2003) was 225%. Likewise, the maximal life-span extension of HBECs in the study by Gu et al. (2003) was 257%. Similarly, in the present study, we observed a maximal increase in cell lifespan of 230% (in microglia cultured in the absence of continual CSF stimulation) and 375% (in microglia cultured in continual C SF stimulation). Why are rat microglia not immortalized, and only have an extension in life-span, following retroviral telomerase transduction? Perhaps the ability of exogenous telomerase to act on telomeres and extend cell life-span is affected by the promoter stre ngth, site of integration within the genome of the host cell, levels of endogenous rTR (rat telomerase RNA component) and rTRT (rat telomerase reverse transcriptase) presen t, telomere and/or telomerase-associated proteins, and the levels of telomerase activity produced. A threshold level of telomerase activity may be required for immortalization, as previous studies have shown that transduced cells exhibiting low levels of telo merase are insufficient to prevent telomere shortening (Bodnar et al., 1998). In the presen t study, lack of immort ality suggests that hTRT gene was not incorporated into the genome, but was rather only transiently expressed and was maintained as an extrachromosomal plasmid which was eventually

PAGE 135

122 lost/degraded over time. In addition, telomera se activity may seem apparent when total protein is isolated and measured following in vitro culturing, yet the enzyme could be inhibited by a repressor molecule while in vitro or in vivo which may or may not be present within the total protein pool during analysis. Additional in vitro or in vivo molecules may also play critical roles in regulating telomerase activity. Thus, the pattern of telomerase activity, as determined by in vitro total protein analysis, may not correlate precisely with or imply telomere maintenance in vitro or in vivo (Ouellette et al., 1999). The level of telomerase expression/activity required to adequately maintain telomere length in any cell type has not been determ ined, and thus differe nt expression/activity patterns may exist when comparing in vitro analyses to the actual in vivo environment. The role that microglial telomere shorte ning and senescence plays in normal brain aging and in AD is not understood. Cellular se nescence in microglia has not been studied as a contributing factor in neuron and synapse loss during aging. Since microglia can normally clear A (amyloid-beta peptide) (Frautschy et al., 1992), and are known to proliferate in the adult brain (and are thus susceptible to telomere shortening), they represent an attractive cell population for telome rase immortalization. Previous research in our laboratory has found that telomeres in rat microglia (the only species examined thus far) do shorten with age when cultured in vitro (Flanary and Streit, 2004). However, cells in vivo are exposed to a multitude of cell signals and effector molecules, as well as an extracellular environment that is cu rrently impossible to replicate in an in vitro environment. Thus, while learning from in vitro experimentation will help take us take the research to the next level ( in vivo ), it will be this next le vel that holds the most promise for clinical intervention of any disease. Experiments using telomerase

PAGE 136

123 activation/over-expression may pave the way for establishing an anti-aging therapy that can be applied to an aging in vivo multicellular system (i.e., humans). To date, all research published on telomerase transduction/immortalization in has been performed in vitro Thus, while we may attain remarkable, even unprecedented, success in these in vitro experiments, the in vivo animal studies that are essent ial and will eventually follow may or may not show similar results. Th erefore, placing an em phasis on proceeding to in vivo trials is a necessary second step that to date has yet to be taken. Additional experiments (both in vitro and ex vivo ), as well as human clinical trials, are necessary in order to accurately confirm or deny whether telomerase i nduction/over-expression within microglia in vivo in a telomerase-negative multicellular organism can slow/prevent their telomere shortening and senescence, and as a result, slow/prevent the onset of AD. If microglia undergo telome re shortening with age in vivo in the adult human brain, then this suggests that their subsequent entr y into senescence also occurs. Indeed, future studies in our laboratory will examine telomere length and telomerase activity in human microglia isolated post-mortem from nor mal and AD-demented individuals. As senescent microglial cells amass, they may b ecome dysfunctional and less able to sustain their neuron-supporting functions, ultimately leading to neuronal dysfunction and the eventual death of the neurons they once suppor ted. Neuronal death results in a loss of synaptic connections, which ultimately is th e causal factor of pr ogressive memory loss with age. Thus, preventing telomere shortening (e.g., with exogenous telomerase delivery) in microglia may prevent their senescence and enable these cells to carry out their normal functions for a longer period of time. In vivo re-implantation of ex vivo transduced rat microglia has been successfu lly performed previously (Mordelet et al.,

PAGE 137

124 2002; Watanabe et al., 2002), supporting the not ion that the same could hold true for humans in vivo Perhaps extending the life-span of microglia by immortalizing them with telomerase could enable them to function nor mally for a longer period of time. This may enable these cells to adequately clear A before it aggregates into neurotoxic plaques. If plaque formation can be prevented, then so might neuronal cell death and synapse loss. If synapses can be maintained within the brain for a longer period of time, then additional years or decades of critical thinking and memory retrieval ma y be possible. A concern is that expression of telomerase in normal so matic cells (e.g., microglia) may induce, or aid in the development of, tumorigenesis. Howe ver, other telomerase-transduced cell types have not resulted in the acquis ition of tumorigenic properties (Belair et al., 1997; Jiang et al., 1999; Morales et al., 1999; Harley, 2002). In the curren t study, we have not observed any gross phenotypic or morphol ogical characteristics (e.g., loss of contact inhibition), or anchorage-independent growth of transduced microglia that indicate tumorigenicity, or that may account for the extended cell life-sp an of telomerase-transduced microglia. A better understanding of the molecular mechanis ms of microglial te lomere biology could provide a novel perspective for understandi ng the development and pathogenesis of neurodegenerative diseases, and lead to the development of new drugs designed to enhance microglial cell function and/or to sl ow microglial telomere shortening (e.g., via a telomerase inducer molecule) and senescence as treatments of AD.

PAGE 138

CHAPTER 7 CONCLUSIONS AND IMPLICATIONS Conclusions Collectively, these experiments have fo cused on studying the hypothesis that with aging, microglia undergo telomere shortening both in vitro and in vivo become increasingly dysfunctional, a nd ultimately enter cellular sene scence. The rationale for this hypothesis is based on the fact that microglia undergo cell division in vivo and are thus susceptible to telomere shortening with age. If this situation does indeed occur in vivo in multicellular te lomerase-negative organisms (e.g., humans), it may lead to a decline in microglial cell function with age, which in turn, would inhibit their ability to promote neuronal well-being. Thus, age-related neuron loss may be due to loss of microglial support. Telomere shortening was found to occur in cultured rat microglia in vitro especially under periods of high proliferation (i.e., GM -CSF stimulation), and with additional cell divisions (i.e ., with increasing culture ar ea). Microglia undergoing continual rapid division in vitro apparently use up the majority of their replicative potential prematurely, with the increased proliferation corresponding to both increased telomere attrition and earlier entr y into senescence. An inte resting future study would be to examine telomere length in acutely-isolated [i.e., via fluorescence-activated cell sorting (FACS)] microglia to determine if telomere shortening also occurs in vivo and at what rate. 125

PAGE 139

126 Telomere shortening also occurred with ag e in rat cerebellum and cortex tissues in vivo However, since the tissue samples analyzed represented a mixed population of cell types (both glial and neuronal), it remained difficult to conclusively state that telomere shortening occurred in microglial cells specifically with age in vivo Thus, future experiments will likely focus on examining telomere dynamics in acutely-isolated (i.e., via FACS) microglia to determine if microglia are the cell type responsible for the decrease in telomere length evident with age. In addi tion, looking beyond 6 months of age will provide a more detailed understandi ng of the rate of shortening with age in specific brain tissue types. Perhaps one of th e most important studies to be performed in this area is to examine human microglia l cells and measure telomere length and telomerase activity to determine if they undergo telomere shortening with age and incidence of Alzheimers disease (AD). Co incidentally, our labor atory has currently received five cases (i.e., hu man microglial cell pellets) to date from the Sun Health Research Institute (Sun City, AZ), with additional cases being collected. This experiment, if successful in showing telome re loss with age and incidence of AD in human microglia, would provide the most support to our hypothesis that telomere shortening in microglia contributes to normal brain aging and memory loss. Maintenance and extension of telomere length via the action of the telomerase enzyme was shown to occur in activated, pro liferating microglia that accumulate in the axotomized rat facial motor nuclei. This result was somewhat surprising, since we hypothesized that with increased cell division in vivo that microglia would undergo telomere shortening. However, the fact th at telomerase activity was much higher in axotomized whole facial nuclei tissue suggests that the increase in telomere length was

PAGE 140

127 due to a similar increase in activity. Interestingly, a similar situation occurred in vitro when microglia were stimulated to rapidly divide with GM-C SF, since this also caused dramatic telomere lengthening initially (followed by telomere attrition). It is unknown how telomere length would change at future time-points (i.e., more than four days) following axotomy, however, this will be the su bject of future studies. In addition, since the increase in telomere length was shown to occur in whole facial nucleus tissues, and not in microglia directly, additional studies examining telomere length and telomerase activity at various time-points in acutely-i solated (i.e., via FACS) microglia following single and repeated axotomy are warranted to confirm that the increases evident in telomere length in whole facial nuclei tissues stem from an increase microglial telomere length. If repeated axotomy does cause telo mere shortening in microglia, this would have implications and a correlation to repeat ed head injury cases in humans, and would suggest that head injuries could cause in in crease in telomere shortening and senescence (due to increased division rate of microglia), leading to premature memory loss and dementia, which is known to occur in these i ndividuals (e.g., football and soccer players, boxers). Treatment of rat microglial cells with vitamin E caused dramatic microglial proliferation in vitro This result was confirmed on nu merous instances using different stocks/lots of vitamin E, as well as different cultures of rat microglia. The increase in proliferation evident also resulted in a concomitant decrease in telomere length and telomerase activity. Cultures treated in parallel with the antioxidant -lipoic acid showed no proliferative response, indica ting that the mitogenic effect of vitamin E is independent of its antioxidative action. Vitamin E-induced rat microg lial proliferation was higher

PAGE 141

128 than in cells stimulated with GM-CSF (a well-known and characterized mitogen), suggesting that vitamin E is the most pot ent known microglial mitogen. Vitamin Etreated microglia undergoing rapid division appa rently use up the majority of their replicative potential prematurely, as evidenced by an increased rate of telomere attrition and earlier entry into cellular senescence, co mpared to controls. Future studies will examine the effects of vitamin E supplementation to determine if telomere shortening occurs in vivo If so, this would have implications for the use of vitamin E as a dietary supplement. While vitamin E is clearly beneficial for health as determined by a number of studies and clinical trials, and irrelevant of whet her telomere shortening occurs in vivo in vitamin E-supplemented rats (especiall y since rats normally express telomerase activity in their microglia), if vitamin E cause s proliferation of microglia in humans, this may contribute to an advanced rate of telomere shortening, early entrance into senescence, and premature development of age-related memory loss (if senescent microglia are truly a contributing factor). Retroviral delivery of the protein component of telomerase appears to extend cell life-span and delay senescence in both rat microglia and astrocytes. However, only a GFP vector was shown to have been successfu lly transduced in parallel experiments, which suggests the successful tr ansduction of the telomerase vector, but begs the question of whether or not the telomera se vector actually entered the cells. Current studies in our laboratory are attempting to answer this very question using PCR-based methodologies to detect the presence of the vector DNA and relati ve copy number of the telomerase gene). Implications If telomere shortening occurs in human mi croglia with age and incidence of head injury and dementia, then it seems reasonable that re-lengthening telomeres and/or

PAGE 142

129 preventing further telomere erosion may prove beneficial as a potential treatment to the dysfunction and senescence that may normally with age in microglia. If telomere shortening occurs in human microglia in vivo this may lead to a decline in cell function with age and inhibit their ability promote neuronal well-being. Thus, understanding the molecular mechanisms of microglial telome re biology, how and why telomere shortening triggers cellular senescence, and how microglia are involved in agerelated deterioration of neuronal function could provide a novel perspective to further understand the normal aging process in microglia as well as th e origins of CNS pathogenesis. This understanding could lead to the developm ent of new drugs designed to enhance microglial cell function and/or slow microglia l telomere shortening and senescence as potential treatments for AD in humans. One possible therapy may be to induce telomerase expression/activ ity, either by activating endogenous telomerase via removal of repressor protein(s), or by the delivery of biologically-active and functional telomerase enzyme directly to cells (especially those that are mitotically-active). This treatment may slow/prevent telomere shortening and senescen ce in microglia and enable these cells to promote neuronal well-being and perform othe r vital functions (e.g., clearing amyloid) for a longer time period, which may permit addi tional years of critical thinking ability and memory retrieval, and slow/prevent th e onset of pathological conditions involving senescent microglia

PAGE 143

LIST OF REFERENCES Akeson AL, Woods CW, Mosher LB, Thomas CE, Jackson RL. 1991. Inhibition of IL-1 beta expression in THP-1 cells by probucol and tocopherol. Atherosclerosis. 86, 261-270. Akiyama H, Barger S, Barnum S, Bradt B, Bauer J, Cole GM, Cooper NR, Eikelenboom P, Emmerling M, Fiebich BL, Finch CE, Frautschy S, Griffin WS, Hampel H, Hull M, Landreth G, Lue L, Mrak R, Macken zie IR, McGeer PL, O'Banion MK, Pachter J, Pasinetti G, Plata-Salaman C, Rogers J, Rydel R, Shen Y, Streit W, Strohmeyer R, Tooyoma I, Van Muiswinkel FL, V eerhuis R, Walker D, Webster S, Wegrzyniak B, Wenk G, Wyss-Coray T. 2000. Inflammation and Alzheimers Disease. Neurobiol Aging. 21, 383-421. Allsopp RC, Vaziri H, Patterson C, Goldstei n S, Younglai EV, Futcher AB, Greider CW, Harley CB. 1992. Telomere length pred icts replicative capacity of human fibroblasts. Proc Natl Acad Sci USA. 89, 10114-10118. Allsopp RC, Harley CB. 1995. Evidence for a critical telomere le ngth in senescent human fibroblasts. Exp Cell Res. 219, 130-136. Baek S, Bu Y, Kim H, Kim H. 2004. Telo merase induction in astrocytes of SpragueDawley rat after ischemic brain in jury. Neurosci Lett. 363, 94-96. Baur JA, Zou Y, Shay JW, Wright WE. 2001. Telomere position effect in human cells. Science. 292, 2075-2077. Behl C, Holsboer F. 1998. Oxidative stress in the pathogenesis of Alzheimer's disease and antioxidant neuroprotection. Fo rtschr Neurol Psychiatr. 66, 113-121. Behl C, Moosmann B. 2002. Antioxidant ne uroprotection in Alzheimer's disease as preventive and therapeutic approach Free Radic Biol Med. 33, 182-191. Bekaert S, Koll S, Thas O, Van Oostveldt P. 2002. Comparing telomere length of sister chromatids in human lymphocytes using three-dimensional confocal microscopy. Cytometry. 48, 34-44. Belair CD, Yeager TR, Lopez PM, Rezni koff CA. 1997. Telomerase activity: A biomarker of cell proliferation, not maligna nt transformation. Proc Natl Acad Sci USA. 94, 13677-13682. Bendich A. 1988. Vitamin E and immune functions. Basic Life Sci. 49, 615-620. 130

PAGE 144

131 Bernd A, Batke E, Zahn RK, Muller WE. 1982. Age-dependent gene induction in quail oviduct. XV. Alterations of the poly(A)-associated protein pattern and of the poly(A) chain length of mRNA. Mech Ageing Dev. 19, 361-377. Blasco MA, Rizen M, Greider CW, Hanahan D. 1996. Differential regulation of telomerase activity and telomerase RNA during multi-stage tumorigenesis. Nat Genet. 12, 200-204. Bodnar AG, Ouellette M, Frolkis M, Holt SE Chiu C-P, Morin GB, Harley CB, Shay JW, Lichtsteiner S, Wright WE. 1998. Extension of life-span by introduction of telomerase into normal human cells. Science. 279, 349-352. Bowman PD and Daniel CW. 1975. Characteristics of proliferative cells from young, old, and transformed WI 38 cultures. Adv Exp Med Biol. 53, 107-122. Burger AM, Bibby MC, Double JA. 1997. Telome rase activity in normal and malignant mammalian tissues: Feasibility of telomerase as a target for cancer chemotherapy. Br J Cancer. 75, 516-522. Caporaso GL and Chao MV. 2001. Telomerase and oligodendrocyte differentiation. J Neurobiol. 49, 224-234. Caporaso GL, Lim DA, Alvarez-Buylla A, Ch ao MV. 2003. Telomerase activity in the subventricular zone of adult mice. Mol Cell Neurosci. 23, 693-702. Carson MJ, Reilly CR, Sutcliffe JG, Lo D. 1998. Mature microglia resemble immature antigen-presenting cells. Glia. 22, 72-85. Cech TR, Nakamura TM, Linger J. 1997. Telome rase is a true reverse transcriptase. A review. Biochemistry (Mosc). 62, 1202-1205. Chaudhary G, Sinha K, Gupta YK. 2003. Pr otective effect of exogenous administration of alpha-tocopherol in middle cerebral arte ry occlusion model of cerebral ischemia in rats. Fundam Clin Pharmacol. 17, 703-707. Cherif H, Tarry JL, Ozanne SE, Hales CN. 2003. Ageing and telomeres: A study into organand gender-specific telomere s hortening. Nucleic Acids Res. 31, 15761583. Clifton GL, Lyeth BG, Jenkins LW, Taft WC DeLorenzo RJ, Hayes RL. 1989. Effect of D, alpha-tocopheryl succinate and polyethylene glycol on performance tests after fluid percussion brain injury. J Neurotrauma. 6, 71-81. Colton CA, Gilbert DL. 1987. Production of superoxide anions by a CNS macrophage, the microglia. FEBS Lett. 223, 284-288.

PAGE 145

132 Conte V, Uryu K, Fujimoto S, Yao Y, Rokach J, Longhi L, Trojanowski JQ, Lee VM, McIntosh TK, Pratico D. 2004. Vitamin E reduces amyloidosis and improves cognitive function in Tg2576 mice following re petitive concussive brain injury. J Neurochem. 90, 758-764. Coviello-McLaughlin GM and Prowse KR. 1997. Telomere length regulation during postnatal development and ageing in Mus spretus Nucleic Acids Res. 25, 30513058. Cui W, Aslam S, Fletcher J, Wylie D, Clinton M, Clark AJ. 2002. Stabilization of telomere length and karyotypic stability ar e directly correlated with the level of hTERT gene expression in primary fibroblasts. J Biol Chem. 277, 38531-38539. de Lange T. 1994. Activation of telomerase in a human tumor. Proc Natl Acad Sci USA. 91, 2882-2885. de Pauw ED, Verwoerd NP, Duinkerken N, Wilemze R, Raap AD, Fibbe WE, Tanke HJ. 1998. Assessment of telomere length in hematopoietic interphase cells using in situ hybridization and digital fluorescence microscopy. Cytometry. 32, 163-169. de Simone R, Giampaolo A, Giometto B, Gallo P, Levi G, Peschle C, Aloisi F. 1995. The costimulatory molecule B7 is expre ssed on human microglia in culture and in multiple sclerosis acute lesions. J Neuropathol Exp Neurol. 54, 175-187. Devaraj S, Li D, Jialal I. 1996. The e ffects of alpha tocopherol supplementation on monocyte function. Decreased lipid oxidation, interleuki n 1 beta secretion, and monocyte adhesion to endothelium J Clin Invest. 98, 756-763. Devaraj S, Jialal I. 1999. Alpha-tocopherol decreases interleukin-1 beta release from activated human monocytes by inhibition of 5-lipoxygenase. Arterioscler Thromb Vasc Biol. 19, 1125-1133. Devi SA, Kiran TR. 2004. Regi onal responses in antioxidant system to exercise training and dietary vitamin E in aging rat brain. Neurobiol Aging. 25, 501-508. Dickson DW, Crystal HA, Mattiace LA, Masur DM, Blau AD, Davies P, Yen SH, Aronson MK. 1992. Identification of normal and pathological aging in prospectively studied nondemented elde rly humans. Neurobiol Aging. 13, 179189. di Donna S, Mamchaoui K, Cooper RN, Seigne urin-Venin S, Tremblay J, Butler-Browne GS, Mouly V. 2003. Telomerase can exte nd the proliferative capacity of human myoblasts, but does not lead to their immo rtalization. Mol Cancer Res. 1, 643653.

PAGE 146

133 Dimri GP, Lee X, Basile G, Acosta M, Sco tt G, Roskelley C, Medrano EE, Linskens M, Rubel I, Pereira-Smith O, Peacocke M, Campisi J. 1995. A biomarker that identifies senescent human cells in culture and in aging skin in vivo Proc Natl Acad Sci USA. 92, 9363-9367. Egger T, Hammer A, Wintersperger A, Goti D, Malle E, Sattler W. 2001. Modulation of microglial superoxide production by alpha-tocopherol in vitro : Attenuation of p67(phox) translocation by a protein phosphatase-dependent pathway. J Neurochem. 79, 1169-1182. Egger T, Schuligoi R, Wintersperger A, Aman n R, Malle E, Sattler W. 2003. Vitamin E (alpha-tocopherol) attenuat es cyclo-oxygenase 2 transc ription and synthesis in immortalized murine BV-2 mi croglia. Biochem J. 370, 459-467. Eglitis MA, Mezey E. 1997. Hematopoietic ce lls differentiate into both microglia and macroglia in the brains of adult mice. Proc Natl Acad Sci USA. 94, 4080-4085. Flanary B. 2004. Book Review: Cells, aging, and human disease. Rejuvenation Res. 7(2), 146-147. Flanary BE, Streit WJ. 2003. Telomeres shorten with age in rat cerebellum and cortex in vivo J Anti-Aging Med. 6(4), 299-308. Flanary BE, Streit WJ. 2004. Progressive telomere shortening occurs in cultured rat microglia, but not astrocytes. Glia. 45, 75-88. Flanary BE, Streit WJ. Axotomy increases telomere length, telomerase activity and protein in axotomy-activated microglia. Submitted for review to Neurobiol Aging. Ford AL, Goodsall AL, Hickey WF, Sedgw ick JD. 1995. Normal adult ramified microglia separated from other central nervous system macrophages by flow cytometric sorting. Phenotypic differences defined and direct ex vivo antigen presentation to myelin basic protein-reactive CD4+ T cells compared. J Immunol. 154, 4309-4321. Fossel M. 1998. Telomerase and the aging cel l: Implications for human health. JAMA. 279, 1732-1735. Fossel M. 2000. Cell senescence in human agi ng: A review of the theory. In vivo. 14, 29-34. Fossel M. 2004. Cells, Aging, and Human Di sease. Oxford University Press, New York, NY. 489 p. Frautschy SA, Cole GM, Baird A. 1992. Ph agocytosis and depos ition of vascular amyloid in rat brains injected with Alzheimer -amyloid. Am J Pathol. 140, 13891399.

PAGE 147

134 Fu W, Killen M, Culmsee C, Dhar S, Pandita TK, Mattson MP. 2000. The catalytic subunit of telomerase is e xpressed in developing brain neurons and serves a cell survival-promoting function. J Mol Neurosci. 14, 3-15. Fu W, Lee J, Guo Z, Mattson MP. 2002. Seiz ures and tissue injury induce telomerase in hippocampal microglial cells. Exp Neurol. 178, 294-300. Funk WD, Wang CK, Shelton DN, Harley CB, Pagon GD, Hoeffler WK. 2000. Telomerase expression restores dermal integrity to in vitro -aged fibroblasts in a reconstituted skin model. Exp Cell Res. 258, 270-278. Giulian D, Ingeman JE. 1988. Colony-stimul ating factors as promoters of ameboid microglia. J Neurosci. 8, 4707-4717. Godbout JP, Berg BM, Kelley KW, Johnson RW. 2004. Alpha-tocopherol reduces lipopolysaccharide-induced per oxide radical formation a nd interleukin-6 secretion in primary murine microglia and in brain. J Neuroimmunol. 149, 101-109. Golubovskaya VM, Presnell SC, Hooth MJ, Smith GJ, Kaufmann WK. 1997. Expression of telomerase in normal and mali gnant rat hepatic epithelia. Oncogene. 4, 1233-1240. Gonzalez R, Sanchez de Medina F, Galvez J, Rodriguez-Cabezas ME, Duarte J, Zarzuelo A. 2001. Dietary vitamin E supplementati on protects the rat la rge intestine from experimental inflammation. In t J Vitam Nutr Res. 71, 243-250. Gonzalez-Perez O, Gonzalez-Castaneda RE Huerta M, Luquin S, Gomez-Pinedo U, Sanchez-Almaraz E, Navarro-Ruiz A, Garcia -Estrada J. 2002. Beneficial effects of alpha-lipoic acid vitamin E on neurological deficit, r eactive gliosis and neuronal remodeling in the prenumbra of the ischem ic rat brain. Neurosci Lett. 321, 100104. Gottschling DE, Cech TR. 1984. Chromatin structure of the molecular ends of Oxytricha macronuclear DNA: Phased nuc leosomes and a telomeric complex. Cell. 38, 501-510. Graeber MB, Kreutzberg GW. 1986. Astrocytes increase in glial fibr illary acidic protein during retrograde changes of facial mo tor neurons. J Neurocytol. 15, 363-373. Graeber MB, Tetzlaff W, Streit WJ, Kreutz berg GW. 1988. Micr oglial cells but not astrocytes undergo mitosis following faci al nerve axotomy. Neurosci Lett. 85, 317-321. Grammas P, Hamdheydari L, Benaksas EJ Mou S, Pye QN, Wechter WJ, Floyd RA, Stewart C, Hensley K. 2004. Anti-inflamma tory effects of tocopherol metabolites. Biochem Biophys Res Commun. 319, 1047-1052.

PAGE 148

135 Grant JD, Broccoli D, Muquit M, Manion FJ, Tisdall J, Ochs MF. 2001. Telometric: A tool providing simplified, reproducible measurements of telomeric DNA from constant field agarose gels. Biotechniques. 31, 1314-1316. Greider CW. 1991. Telomerase is pr ocessive. Mol Cell Biol. 11, 4572-4580. Greider CW, Blackburn EH. 1985. Identificat ion of a specific telomere terminal transferase enzyme with tw o kinds of primer specificity. Cell. 51, 405-413. Grundman M, Delaney P. 2002. Antioxidant strategies for Al zheimers disease. Proc Nutr Soc. 61, 191-202. Gu X, Zhang J, Brann DW, Yu FS. 2003. Br ain and retinal vascul ar endothelial cells with extended life span established by ect opic expression of telomerase. Invest Ophthalmol Vis Sci. 44, 3219-3225. Guo W, Kang MK, Kim HJ, Park NH. 1998. Immortalization of human oral keratinocytes is associated with elevation of telomerase activity and shortening of telomere length. Oncol Rep. 5, 799-804. Halliwell B, Gutteridge JMC. 1985. Oxygen ra dicals in the nervous system. Trends Neurosci. 8, 22-26. Halvorsen TL, Leibowitz G, Levine F. 1999. Telomerase activity is sufficient to allow transformed cells to escape from crisis. Mol Cell Biol. 19, 1864-1870. Haramaki N, Packer L, Assadnazari H, Zimm er G. 1993. Cardiac recovery during postischemic reperfusion is improved by combin ation of vitamin E with dihydrolipoic acid. Biochem Biophys Res Comm. 196, 1101-1107. Harley CB. 1991. Telomere loss: Mitotic cl ock or genetic time bomb? Mutat Res. 256, 271-282. Harley CB, Futcher B, Greider CW. 1990. Telomeres shorten during ageing of human fibroblasts. Nature. 345, 458-460. Harley CB, Vaziri H, Counter CM, Alls opp RC. 1992. The telomere hypothesis of cellular aging. Exp Gerontol. 27, 375-382. Harley CB. 2002. Telomerase is not an oncogene. Oncogene. 21, 494-502. Hayflick L. 1961. The serial cultivation of human diploid cell strains. Exp Cell Res. 25, 585-621. Hayflick L. 1965. The limited in vitro lifetime of human diploid cell strains. Exp Cell Res. 37, 614-636.

PAGE 149

136 Hemann MT, Strong MA, Hao L-Y, Greider CW. 2001. The shortest telomere, not average telomere length, is critical for cell viability and chromosome stability. Cell. 107, 67-77. Heppner FL, Roth K, Nitsch R, Hailer NP. 1998. Vitamin E induces ramification and downregulation of adhesion molecules in cultured microglial cells. Glia. 22,180188. Hess DC, Abe T, Hill WD, Studdard AM, Carothers J, Masuya M, Fleming PA, Drake CJ, Ogawa M. 2004. Hematopoietic origin of microglial and perivascular cells in brain. Exp Neurol. 186, 134-144. Hiyama K, Hirai Y, Kyoizumi S, Akiyam a M, Hiyama E, Piatyszek MA, Shay JW, Ishioka S, Yamakido M. 1995. Activation of telomerase in human lymphocytes and hematopoietic progenitor cells. J Immunol. 155, 3711-3715. Hooijberg E, Ruizendaal JJ, Snijders PJ, Ku eter EW, Walboomers JM, Spits H. 2000. Immortalization of human CD8+ T cell cl ones by ectopic expression of telomerase reverse transcriptase. J Immunol. 165, 4239-4245. Hurley SD, Coleman PD. 2003. Facial nerv e axotomy in aged and young adult rats: Analysis of the glial respons e. Neurobiol Aging. 24, 511-518. Ikeda Y, Mochizuki Y, Nakamura Y, Dohi K, Matsumoto H, Jimbo H, Hayashi M, Matsumoto K, Yoshikawa T, Murase H, Sato K. 2000. Protective effect of a novel vitamin E derivative on experimental trauma tic brain edema in rats preliminary study. Acta Neurochir Suppl. 76, 343-345. Ikeda K, Negishi H, Yamori Y. 2003. Anti oxidant nutrients and hypoxia/ischemia brain injury in rodents. Toxicology. 189, 55-61. Inci S, Ozcan OE, Kilinc K. 1998. Time-level relationship for lipid peroxidation and the protective effect of alpha-toc opherol in experimental mild and severe brain injury. Neurosurgery. 43, 330-335. Itagaki S, McGeer PL, Akiyama H, Zhu S, Selkoe D. 1989. Relationship of microglia and astrocytes to amyloid deposits of Al zheimers disease. J Neuroimmunol. 24, 173-182. Jackson CV, Holland AJ, Williams CA, Dickerson JW. 1988. Vitamin E and Alzheimers disease in subj ects with Downs syndrome. J Ment Defic Res. 32, 479-484. Jellinger KA, Stadelmann CH. 2000. The en igma of cell death in neurodegenerative disorders. J Neural Transm Suppl. 60, 21-36.

PAGE 150

137 Jiang XR, Jimenez G, Chang E, Frolkis M, Kusler B, Sage M, Beeche M, Bodnar AG, Wahl GM, Tlsty TD, Chiu CP. 1999. Telo merase expression in human somatic cells does not induce changes associated with a transformed phenotype. Nat Gen. 21, 111-114. Joseph JA, Shukitt-Hale B, Denisova NA, Prior RL, Cao G, Martin A, Taglialatela G, Bickford PC. 1998. Long-term dietar y strawberry, spinach, or vitamin E supplementation retards the onset of age -related neuronal signal-transduction and cognitive behavioral deficits. J Neurosci. 18, 8047-8055. Joseph JA, Shukitt-Hale B, Denisova NA, Bielinski D, Martin A, McEwen JJ, Bickford PC. 1999. Reversals of age-related dec lines in neuronal signal transduction, cognitive, and motor behavioral deficits with blueberry, spinach, or strawberry dietary supplementation. J Neurosci. 19, 8114-8121. Kaempf-Rotzoll DE, Traber MG, Arai H. 2003. Vitamin E and transfer proteins. Curr Opin Lipidol. 14, 249-254. Kajstura J, Pertoldi B, Leri A, Beltrami CA, Deptala A, Darzynkiewicz Z, Anversa P. 2000. Telomere shortening is an in vivo marker of myocyte re plication and aging. Am J Pathol. 156, 813-819. Kang HJ, Choi YS, Hong SB, Kim KW, Woo RS, Won SJ, Kim EJ, Jeon HK, Jo SY, Kim TK, Bachoo R, Reynolds IJ, Gwag BJ, Lee HW. 2004. Ectopic expression of the catalytic subunit of telomerase protects against brain injury resulting from ischemia and NMDA-induced neurot oxicity. J Neurosci. 24, 1280-1287. Kawas C, Gray S, Brookmeyer R, Fozard J, Zonderman A. 2000. Age-specific incidence rates of Alzheimers disease: The Baltimore l ongitudinal study of aging. Neurology. 54, 2072-2077. Kim NW, Piatyszek MA, Prowse KR, Harley CB, West MD, Ho PL, Coviello GM, Wright WE, Weinrich SL, Shay JW. 1994. Specific association of human telomerase activity with immortal cells and cancer. Science. 266, 2011-2015. Kim NW, Wu F. 1997. Advances in quantif ication and characterization of telomerase activity by the telomeric repeat amplificat ion protocol (TRAP). Nuc Acids Res. 25, 2595-2597. Kipling D. 1995. The Telomere. Oxford University Press, New York, NY. 224 p. Kipling D, Cooke HJ. 1990. Hypervariable ul tra-long telomeres in mice. Nature. 347, 400-402. Kitabchi AE, Wimalasena J, Barker J. 1980. Specific receptor sites for alpha-tocopherol in purified isolated adrenocortical cell membrane. Biochem Biophys Res Commun. 96, 1739-1746.

PAGE 151

138 Kitabchi AE, Wimalasena J. 1982. Demonstration of specific binding sites for 3H-RRRalpha-tocopherol on human erythrocyt es. Ann NY Acad Sci. 393, 300-314. Klapper W, Shin T, Mattson MP. 2001. Diffe rential regulation of telomerase activity and TERT expression during brain developmen t in mice. J Neurosci Res. 64, 252260. Kodama S, Mori I, Roy K, Yang Z, Suzuki K, Wantanabe M. 2001. Culture conditiondependent senescence-like growth arrest and immortalizatio n in rodent embryo cells. Radiat Res. 155, 254-262. Kreutzberg GW. 1966. Autoradiographisc he Untersuchung ber die Beteiligung von Gliazellen an der axonalen Reaktion in Faci aliskern der Ratte. Acta Neuropath. (Berl). 7, 149-161. Kreutzberg GW. 1996. Microglia: A sensor for pathological ev ents in the CNS. Trends Neurosci. 19, 312-318. Kruk PA, Balajee AS, Rao KS, Bohr VA. 1996. Telomere reduction and telomerase inactivation during neuronal cell differe ntiation. Biochem Biophys Res Commun. 224, 487-492. Kuhn HG, Dickinson-Anson H, Gage FH. 1996. Neurogenesis in the dentate gyrus of the adult rat: Age-related decrease of neuronal progenitor proliferation. J Neurosci. 16, 2027-2033. Lassmann H, Bancher C, Breitschopf H, Wegiel J, Bobinski M, Jellinger K, Wisniewski HM. 1995. Cell death in Alzheimers di sease evaluated by DNA fragmentation in situ. Acta Neuropathol. 89, 35-41. Li Y, Liu L, Barger SW, Mrak RE, Griffin WS. 2001. Vitamin E suppression of microglial activation is neuroprotect ive. J Neurosci Res. 66, 163-170. Lindsey J, McGill NI, Lindsey LA, Green DK, Cooke HJ. 1991. In vivo loss of telomere repeats with age in humans. Mutat Res. 256, 45-48. Mann DM, Yates PO. 1981. The relationship be tween formation of senile plaques and neurofibrillary tang les and changes in nerve cell metabolism in Alzheimer type dementia. Mech Ageing Dev. 17, 395-401. Martin A, Janigian D, Shukitt-Hale B, Prio r RL, Joseph JA. 1999. Effect of vitamin E intake on levels of vitamins E and C in the central nervous system and peripheral tissues: implications for health r ecommendations. Brain Res. 845, 50-59. Martin A, Cherubini A, Andres-Lacueva C, Paniagua M, Joseph J. 2002. Effects of fruits and vegetables on levels of vi tamins E and C in the brain and their association with cognitive performance. J Nutr Health Aging. 6, 392-404.

PAGE 152

139 Mattson MP. 2000. Emerging neuroprotective strategies for Alzheimer's disease: Dietary restriction, telomera se activation, and stem cell therapy. Exp Gerontol. 35, 489-502. Mattson MP, Klapper W. 2001. Emerging roles for telomerase in neuronal development and apoptosis. J Neuro Res. 63, 1-9. McGeer PL, Itagaki S, Tago H, McGeer EG. 1987. Reactive microglia in patients with senile dementia of the Alzheimers type are positive for the histocompatibility glycoprotein HLA-DR. Ne urosci Lett. 79, 195-200. McGeer PL, McGeer EG. 2001. Inflammation, autotoxicity and Alzheimer disease. Neurobiol Aging. 22, 799-809. Meier R, Tomizaki T, Schulze-Briese C, Baumann U, Stocker A. 2003. The molecular basis of vitamin E retention: Structure of human alpha-tocophero l transfer protein. J Mol Biol. 331, 725-734. Meydani SN, Meydani M, Verdon CP, Shapir o AA, Blumberg JB, Hayes KC. 1986. Vitamin E supplementation suppresses prosta glandin E1(2) synthesis and enhances the immune response of aged mice. Mech Ageing Dev. 34, 191-201. Milgram NW, Zicker SC, Head E, Mugge nburg BA, Murphey H, Ikeda-Douglas CJ, Cotman CW. 2002. Dietary enrichment counteracts age-associated cognitive dysfunction in canines. Neurobiol Aging. 23, 737-745. Miller JS, Gavino VC, Ackerman GA, Sharma HM, Milo GE, Geer JC, Cornwell DG. 1980. Triglycerides, lipid droplets, and ly sosomes in aorta smooth muscle cells during the control of cell pr oliferation with polyunsaturat ed fatty acids and vitamin E. Lab Invest. 42, 495-506. Moneta ME, Gehrmann J, Topper R, Banati RB, Kreutzberg GW. 1993. Cell adhesion molecule expression in the regenerating ra t facial nucleus. J Neuroimmunol. 45, 203-206. Morales CP, Holt SE, Ouellette M, Kaur KJ Yan Y, Wilson KS, White MA, Wright WE, Shay JW. 1999. Absence of cancer-asso ciated changes in human fibroblasts immortalized with telomerase. Nat Gen. 21, 115-118. Mordelet E, Kissa K, Calvo CF, Lebastard M, Milon G, van der Werf S, Vidal C, Charneau P. 2002. Brain engraftment of autologous macrophages transduced with a lentiviral flap vector: An approach to complement brain dysfunctions. Gene Ther. 9, 46-52. Morin GB. 1989. The human telomere terminal transferase enzyme is a ribonucleoprotein that synthesizes TTAGGG repeats. Cell. 59, 521-529.

PAGE 153

140 Moyzis RK, Buckingham JM, Cram LS, Dani M, Deaven LL, Jones MD, Meyne J, Ratliff RL, Wu JR. 1988. A highly conserved repetitive DNA sequence, (TTAGGG)n, present at the telomeres of hu man chromosomes. Proc Natl Acad Sci USA. 85, 6622-6626. Murphy DJ, Mavis RD. 1981. Membrane transf er of alpha-tocopherol. Influence of soluble alpha-tocopherol-bi nding factors from the liver, lung, heart, and brain of the rat. J Biol Chem. 256, 10464-10468. Nakajima K, Honda S, Tohyama Y, Imai Y, K ohsaka S, Kurihara T. 2001. Neurotrophin secretion from cultured microglia J Neurosci Res. 65, 322-331. Niu KC, Lin KC, Yang CY, Lin MT. 2003. Pr otective effects of alpha-tocopherol and mannitol in both circulatory shock and cerebra l ischaemia injury in rat heatstroke. Clin Exp Pharmacol Physiol. 10, 745-751. Noguchi T, Ikeda K, Sasaki Y, Yamamoto J, Seki J, Yamagata K, Nara Y, Hara H, Kakuta H, Yamori Y. 2001. Effects of vitamin E and sesamin on hypertension and cerebral thrombogenesis in stroke-prone spontaneously hypertensive rats. Hypertens Res. 24, 735-742. O'Donnell E, Lynch MA. 1998. Dietary an tioxidant supplementation reverses agerelated neuronal changes. Neurobiol Aging. 19, 461-467. Olovnikov AM. 1971. Principle of margi notomy in template synthesis of polynucleotides. Doklady Akademii Nauk (SSR). 201, 1496-1499. Olovnikov AM. 1973. A theory of marginotomy. The incomplete copying of template margin in enzymatic synthesis of polynucleot ides and biological significance of the phenomenon. J Theor Biol. 41, 181-190. Olovnikov AM. 1996. Telomeres, telomerase, and aging: Origin of the theory. Exp Gerontol. 31, 443-448. Oonishi K, Moriguchi S, Kishino Y. 1995. The role of macrophages in increased mitogen response of rat splenic lymphocytes following in vitro incubation with vitamin E. J Nutr Sci Vitaminol (Tokyo). 41, 445-453. Ostenfeld T, Caldwell MA, Prowse KR, Linskens MH, Jauniaux E, Svendsen CN. 2000. Human neural precursor cells expr ess low levels of telomerase in vitro and show diminishing cell proliferation with extensive axonal outgrowth following transplantation. Exp Neurol. 164, 215-226. Ouellette MM, Aisner DL, Savre-Train I, Wright WE, Shay JW. 1999. Telomerase activity does not always imply telomere maintenance. Biochem Biophys Res Comm. 254, 795-803.

PAGE 154

141 Ouellette MM, Liao M, Herbert B-S, Johns on M, Holt SE, Liss HS, Shay JW, Wright WE. 2000. Subsenescent telomere lengths in fibroblasts immortalized by limiting amounts of telomerase. J Biol Chem. 275, 10072-10076. Packer L, Witt EH, Tritschler HJ. 1995. Alpha-lipoic acid as a biological antioxidant. Free Radic Biol Med. 19, 227-250. Pathania V, Syal N, Pathak CM, Khanduja KL. 1999. Vitamin E suppresses the induction of reactive oxygen species re lease by lipopolysacchar ide, interleukin1beta and tumor necrosis factor-alpha in rat alveolar macrophages. J Nutr Sci Vitaminol. 45, 675-686. Payne CM, Glasser L, Tischler ME, Wyckoff D, Cromey D, Fiederlein R, Bohnert O. 1994. Programmed cell death of th e normal human neutrophil: An in vitro model of senescence. Microsc Res Tech. 28, 327-344. Perrig WJ, Perrig P, Stahelin HB. 1997. The relation between antioxidants and memory performance in the old and very ol d. J Am Geriatr Soc. 45, 718-724. Poon SS, Martens UM, Ward RK, Lansdorp PM 1999. Telomere length measurements using digital fluorescence microscopy. Cytometry. 36, 267-278. Prokof'eva VV, Pleskach NM, Bozhkov VM, Demin VG, Liashko VN. 1982. Cellular DNA repair, proliferative act ivity and biochemical char acteristics in the human premature aging syndrome progeria. Tsitologiia. 24, 592-603. Prowse KR, Greider CW. 1995. Developmenta l and tissue-specific regulation of mouse telomerase and telomere length. Pr oc Natl Acad Sci USA. 92, 4818-4822. Raivich G, Gehrmann J, Kreutzberg GW 1991. Increase of macrophage colonystimulating factor and granulocyte-macropha ge colony-stimulating factor receptors in the regenerating rat facial nucl eus. J Neurosci Res. 30, 682-686. Raivich G, Moreno-Flores MT, Moller JC Kreutzberg GW. 1994. Inhibition of posttraumatic microglial proliferation in a genetic model of macrophage colonystimulating factor deficiency in th e mouse. Eur J Neurosci. 6, 1615-1618. Raivich G, Jones LL, Kloss CU, Werner A, Neumann H, Kreutzberg GW. 1998. Immune surveillance in the injured nervous system: T-lymphocytes invade the axotomized mouse facial motor nucleus and aggregate around sites of neuronal degeneration. J Neurosci. 18, 5804-5816. Rogers J, Luber-Narod J, Styren SD, Civi n WH. 1988. Expression of immune systemassociated antigens by cells of the human central nervous system: Relationship to the pathology of Alzheimers dis ease. Neurobiol Aging. 9, 339-349.

PAGE 155

142 Roy RM, Petrella M, Ross WM. 1991. Modifi cation of mitogen-induc ed proliferation of murine splenic lymphocytes by in vitro tocopherol. Immunopharmacol Immunotoxicol. 13, 531-550. Sakai S, Moriguchi S. 1997. Long-term f eeding of high vitamin E diet improves the decreased mitogen response of rat splenic lymphocytes with aging. J Nutr Sci Vitaminol (Tokyo). 43, 113-122. Sano M, Ernesto C, Thomas RG, Klauber MR, Schafer K, Grundman M, Woodbury P, Growdon J, Cotman CW, Pfeiffer E, Schneid er LS, Thal LJ. 1997. A controlled trial of selegiline, alpha-tocopherol, or both as treatment for Alzheimer's disease. N Engl J Med. 336, 1216-1222. Schiefer J, Kampe K, Dodt HU, Zieglgansb erger W, Kreutzberg GW. 1999. Microglial motility in the rat facial nucleus followi ng peripheral axotomy. J Neurocytol. 28, 439-453. Scholich H, Murphy ME, Sies H. 1989. Anti oxidant activity of dihydrolipoate against microsomal lipid peroxidation and its dependence on -tocopherol. Biochem Ciophys Acta. 1101, 256-261. Sedgwick JD, Schwender S, Imrich H, Dorri es R, Butcher GW, ter Meulen V. 1991. Isolation and direct charact erization of resident micr oglial cells from the normal and inflamed central nervous system. Proc Natl Acad Sci USA. 88, 7438-7442. Seigneurin-Venin S, Bernard V, Tremblay JP. 2000. Telomerase allows the immortalization of T antigen-positive DMD myoblasts: A new source of cells for gene transfer application. Gene Ther. 7, 619-623. Shay JW. 1999. At the end of the millennium, a view of the end. Nat Genet. 23, 382383. Shay JW, Pereira-Smith OM, Wright WE. 1991. A role for both RB and p53 in the regulation of human cellular sene scence. Exp Cell Res. 196, 33-39. Shelton DN, Chang E, Whittier PS, Choi D, Funk WD. 1999. Microarray analysis of replicative senescence. Curr Biol. 9, 939-945. Sheng JG, Mrak RE, Griffin WS. 1998. Enlarged and phagocytic, but not primed, interleukin-1 alpha-immunoreactive microglia increase with age in normal human brain. Acta Neuropathol. 95, 229-234. Socci DJ, Crandall BM, Arendash GW. 1995. Chronic antioxidant treatment improves the cognitive performance of aged rats. Brain Res. 693, 88-94. Spaulding C, Guo W, Effros RB. 1999. Resi stance to apoptosis in human CD8+ T cells that reach replicative senescence afte r multiple rounds of antigen-specific proliferation. Exp Gerontol. 34, 633-644.

PAGE 156

143 Steward O. 2000. Functional Neuroscience Springer-Verlag, New York, NY. p15. Stohrer M, Eichinger A, Schl achter M, Stangassinger M. 1998. Protective effect of vitamin E in a rat model of focal cereb ral ischemia. Z Naturforsch. 53, 273-278. Stoyanovsky DA, Goldman R, Darrow RM Organisciak DT, Kagan VE. 1995. Endogenous ascorbate regenerates vitamin E in the retina directly and in combination with exogenous dihydroli poic acid. Curr Eye Res. 14, 181-189. Streit WJ. 1990. An improved staining met hod for rat microglial cells using the lectin from Griffonia simplicifolia (GSA I-B4). J Histochem Cytochem. 38, 1683-1686. Streit WJ. 1996. The role of microglia in brain injury. Neurotoxicology. 17, 671-678. Streit WJ. 2002a. Physiology and pathophys iology of microglial cell function. In: Streit WJ, editor. Microglia in the re generating and degenerating CNS. Springer Verlag, New York, NY. p 1-14. Streit WJ. 2002b. Microglia as neuroprotec tive, immunocompetent cells of the CNS. Glia. 40, 133-139. Streit WJ, Kreutzberg GW. 1988. Response of endogenous glial cells to motor neuron degeneration induced by toxic rici n. J Comp Neurol. 268, 248-263. Streit WJ, Graeber MB, Kreutzberg GW 1989. Peripheral nerve lesion produces increased levels of major histocompatibility complex antigens in the central nervous system. J Neuroimmunol. 21, 117-123. Streit WJ, Kincaid-Colton CA. 1995. The brains immune system. Sci Am. 273, 58-61. Streit WJ, Sparks DL. 1997. Activation of microglia in the brains of humans with heart disease and hypercholesterolemic rabbits. J Mol Med. 75, 130-138. Streit WJ, Walter SA, Pennell NA. 1999. Reac tive microgliosis. Progress Neurobiol. 57, 563-581. Streit WJ, Hurley SD, McGraw TS, Semple -Rowland SL. 2000. Comparative evaluation of cytokine profiles and reac tive gliosis supports a critical role fo r interleukin-6 in neuron-glia signaling duri ng regeneration. J Neurosci Res. 61, 10-20. Streit WJ, Mrak RE, Griffi n WS. 2004a. Microglia and neuroinflammation: A pathological perspective. J Neuroinflammation. 1, 14. Streit WJ, Sammons NW, Kuhns AJ, Sparks DL. 2004b. Dystrophic microglia in the aging human brain. Glia. 45, 208-212. Suzumura A, Sawada M, Yamamoto H, Marunouchi T. 1990. Effects of colony stimulating factors on isolated microglia in vitro J Neuroimmunol. 30, 111-120.

PAGE 157

144 Svensson M, Mattsson P, Aldskogius H. 1994. A bromodeoxyuridine labelling study of proliferating cells in the brainstem fo llowing hypoglossal nerve transection. J Anat. 185, 537-542. Szyper-Kravitz M, Uziel O, Shapiro H, Radna y J, Katz T, Rowe JM, Lishner M, Lahav M. 2003. Granulocyte colony-stimulati ng factor administration upregulates telomerase activity in CD34+ haematopoi etic cells and may prevent telomere attrition after chemotherapy. Br J Haematol. 120, 329-336. Tagami M, Ikeda K, Yamagata K, Nara Y, Fujino H, Kubota A, Numano F, Yamori Y. 1999. Vitamin E prevents apoptosis in hippocampal neurons caused by cerebral ischemia and reperfusion in stroke-prone spontaneously hypertensive rats. Lab Invest. 79, 609-615. Tengerdy RP. 1989. Vitamin E, immune response, and disease resistance. Ann NY Acad Sci. 570, 335-344. Thomas WE. 1992. Brain macrophages: Eval uation of microglia a nd their functions. Brain Res Brain Res Rev. 17, 61-74. Thomas M, Yang L, Hornsby PJ. 2000. Formation of functional tissue from transplanted adrenocortical cells expressing telomerase reverse transcriptase. Nat Biotechnol. 18, 39-42. van der Worp HB, Bar PR, Kappelle LJ, de Wildt DJ. 1998. Dietary vitamin E levels affect outcome of permanent focal cerebr al ischemia in rats. Stroke. 29, 10021005. Vaziri H, Benchimol S. 1998. Reconstitution of telomerase activity in normal human cells leads to elongation of telomeres and extended replicative lif e span. Curr Biol. 8, 279-282. von Zglinicki T. 2002. Oxidative stress shortens telomeres. Trends Biochem Sci. 27, 339-344. Watanabe R, Takase-Yoden S, Fukumitsu H, Nakajima K. 2002. Cell transplantation to the brain with microglia labeled by neuropa thogenic retroviral ve ctor system. Cell Transplant. 11, 471-473. Watson JD. 1972. Origin of concatemer ic T7 DNA. Nat New Biol. 239, 197-201. Wood JG, Sinclair DA. 2002. TPE or not TPE? Its no longer a question. Trends Pharmacol Sci. 23, 1-4. Wright WE, Pereira-Smith OM, Shay JW. 1989. Reversible ce llular senescence: Implications for immortalization of norm al human diploid fibroblasts. Mol Cell Biol. 9, 3088-3092.

PAGE 158

145 Wright WE, Shay JW. 1992. Telomere posi tional effects and the regulation of cellular senescence. Trends Genet. 8, 193-197. Wright WE, Brasiskyte D, Piatyszek MA, Sh ay JW. 1996a. Experimental elongation of telomeres extends the lifespan of immortal x normal cell hybrids. EMBO J. 15, 1734-1741. Wright WE, Piatyszek MA, Rainey WE, Byrd W, Shay JW. 1996b. Telomerase activity in human germline and embryonic tissues and cells. Dev Genet. 18, 173-179. Wright WE, Shay JW. 2002. Historical claims and current interpreta tions of replicative aging. Nat Biotechnol. 20, 682-688. Yang F, Sun X, Beech W, Teter B, Wu S, Sigel J, Vinters HV, Frautschy SA, Cole GM. 1998. Antibody to caspase-cleaved actin detects apoptosis in differentiated neuroblastoma and plaque-associated ne urons and microglia in Alzheimers disease. Am J Pathol. 152, 379-389. Yang J, Chang E, Cherry AM, Bangs CD, Oei Y, Bodnar A, Bronstein A, Chiu CP, Herron GS. 1999. Human endothelial cell li fe extension by telomerase expression. J Biol Chem. 274, 26141-26148. Yoshida S, Busto R, Abe K, Santiso M, Ginsberg MD. 1985. Compression-induced brain edema in rats: Effect of dietary vitamin E on membrane damage in the brain. Neurology. 35, 126-130. Yudoh K, Matsuno H, Nakazawa F, Katayama R, Kimura T. 2001. Reconstituting telomerase activity using the telomerase catalytic subunit prev ents the telomere shorting and replicative senescence in hu man osteoblasts. J Bone Miner Res. 16, 1453-1464.

PAGE 159

BIOGRAPHICAL SKETCH Barry Eric Flanary began pursuing his Doctor of Philosophy (Ph.D.) degree in biomedical sciences at the University of Fl orida (UFL) College of Medicine Department of Neuroscience in Gainesville in August of 2001. He received his Associate of Science (A.S.) degree in 1996 from Illinois Valley Community College, and his Bachelor of Science (B.S.) (1999) and Master of Science (M.S.) (2001) degrees in biological sciences from Illinois State University (ISU). Since 2000, he has se rved on the Editorial Board and written gerontology literat ure and book reviews for the Journal of Anti-Aging Medicine, the only peer-reviewed scie ntific journal dedicated to publishing research on altering clinical aging and ag e-related diseases. He is currently a member of the American Aging Association, the Gerontologica l Society of America, and the Phi Sigma Biological honor society. During work on his A.S. degree, he received two scholarships, and while working towards his B.S. degree, he was an undergraduate research fellow in a National Science Foundation research traini ng program (collaborative research at undergraduate institutions) under a grant recei ved by ISU. While working on his M.S. thesis, Molecular Cloning, Character ization, and Mutagenesis of the msbB Gene, a Secondary Lipid A Acyltransferase, in Haemophilus parainfluenzae , he received one teaching fellowship, one research fellowsh ip, and two graduate student association research grants. He also served as a Ph i Sigma biological honor society grant review committee member and as a graduate teaching assistant in introductory biology for one year, and introductory microbiology for one se mester at ISU. Fr om 1999 to 2001, he had 146

PAGE 160

147 ongoing involvement as chief scientific advisor for a telomere-based art project (featured in the Siggraph 2001 art gallery) at Art to the Nth Power In c. (www.artn.com/telomeres), a collaborative art group and media lab based in Chicago, IL. During work on his Ph.D. dissertation, Analysis of Rat Microglial Cellular Senescence as Determined by Measurements of Telomere Length and Telo merase Activity, he was competitively selected for and attended a National Institu te on Aging Technical Assistance Workshop for Emerging Scientists and Students Seeking Careers in Aging Research in Boston, MA, twice served as a graduate student mentor, co-authored three research grants received from the Evelyn F. McKnight Brain Resear ch Foundation, co-authored one R01 research grant received from the National Institute on Aging, received two graduate student council grants, one department of Neurosci ence grant, one research fellowship, one graduate fellowship for outstanding resear ch, three fellowships from the American Foundation for Aging Research, one fello wship from the Neurobiology of Aging program, one Science program for excellen ce in science award from the American Association for the Advancem ent of Science, two medical guild research incentive awards, and three endowments from the Br yan W. Robinson Memorial Endowment for the Neurosciences of the Tallahassee Memori al Hospital Foundati on, Inc., at Florida State University. He has presented both his undergraduate and graduate research at numerous local, regional, national, and inte rnational research symposia since 1997. In March of 2004, he presented his research in a platform presentation as a plenary panel member on cellular aging and clin ical interventions at the In augural (first) International Conference on Longevity in Sydney, Australia. His Ph.D. dissertation research focuses on telomere dynamics and cellular senescence in microglia. To date (March, 2005), his

PAGE 161

148 publications include five firstauthor original research articles (3 are published, 2 are submitted for publication), 21 literature and book reviews, 14 abstracts, one book chapter, and one poem. While working on his Ph.D. di ssertation, he married his beautiful wife, Allison.


Permanent Link: http://ufdc.ufl.edu/UFE0009641/00001

Material Information

Title: Analysis of Rat Microglial Cellular Senescence as Determined by Measurements of Telomere Length and Telomerase Activity
Physical Description: Mixed Material
Language: English
Creator: Flanary, Barry Eric ( Dissertant )
Wolfgang J. Striet. ( Thesis advisor )
Shaw, Gerry ( Reviewer )
Narayan,Satya ( Reviewer )
Harrison, Jeff ( Reviewer )
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2005
Copyright Date: 2005

Subjects

Subjects / Keywords: Department of Neuroscience thesis, Ph.D
Aging   ( mesh )
Astrocytes   ( mesh )
Axotomy   ( mesh )
Cell Proliferation   ( mesh )
Microglia   ( mesh )
Research   ( mesh )
Telomerase   ( mesh )
Tocopherols   ( mesh )
Transduction, Genetic   ( mesh )
Dissertations, Academic -- UF -- College of Medicine -- Department of Neuroscience
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
theses   ( marcgt )

Notes

Abstract: Normal somatic cells have a finite replicative capacity, and with each cell division, telomeres (the physical ends of chromosomes) progressively shorten until they reach a critical length, at which point the cells enter cellular senescence. Some cells maintain their telomeres by the action of the telomerase enzyme. Microglia, a non-neuronal cell type residing within the central nervous system (CNS), play vital roles in maintaining neuronal function, health, and survival in both the normal and pathological CNS. Microglia are the only adult cell type in the CNS that exhibit significant mitotic potential, suggesting that these cells have limited life-spans, may rely on proliferation to replace senescent cells, and are thus susceptible to telomere shortening and subsequent cellular senescence. In our studies, we have found that telomere shortening occurred in cultured rat microglia concomitant with their progression to senescence by 32 days in vitro. Telomere shortening also occurred in vivo in both rat cerebellum and cortex from day 21 to approximately 5 months of age (i.e., the oldest age analyzed). Axotomy-activated microglia from the facial nucleus (FN) maintained telomere length (TL) via increased levels of telomerase activity (TA) during periods of high proliferation in vivo. Microglia isolated directly from the axotomized FN via fluorescence-activated cell sorting exhibited increased TA relative to un-operated controls, suggesting that microglia are the primary cell type responsible for the increased TA observed in whole tissue FN samples. Vitamin E induced a significantly high proliferation rate in cultured rat microglia. This high rate of proliferation resulted in a concomitant decrease in TL, TA, and microglial activation. Microglia retrovirally-transduced with telomerase exhibited an increased maximal life-span (ranging from 230 to 375%), and delayed entry into senescence, relative to controls and empty-vector transduced microglia. Telomerase transduction did not immortalize microglia, although these cells exhibited a normal phenotype, and had telomerase activity/protein present well past the time when all control cells had died. Our findings provide an impetus to further investigate rat microglial telomere dynamics in vivo, especially with age, following axotomy, or vitamin E supplementation, as well as in human microglia with age and incidence of Alzheimer s disease.
Subject: aging, astrocyte, axotomy, brain, life, microglia, proliferation, rat, senescence, telomerase, telomere, tocopherol, transduction
General Note: Title from title page of source document.
General Note: Document formatted into pages; contains 161 pages.
General Note: Includes vita.
Thesis: Thesis (Ph.D.)--University of Florida, 2005.
Bibliography: Includes bibliographical references.
General Note: Text (Electronic thesis) in PDF format.

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: aleph - 003322360
System ID: UFE0009641:00001

Permanent Link: http://ufdc.ufl.edu/UFE0009641/00001

Material Information

Title: Analysis of Rat Microglial Cellular Senescence as Determined by Measurements of Telomere Length and Telomerase Activity
Physical Description: Mixed Material
Language: English
Creator: Flanary, Barry Eric ( Dissertant )
Wolfgang J. Striet. ( Thesis advisor )
Shaw, Gerry ( Reviewer )
Narayan,Satya ( Reviewer )
Harrison, Jeff ( Reviewer )
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2005
Copyright Date: 2005

Subjects

Subjects / Keywords: Department of Neuroscience thesis, Ph.D
Aging   ( mesh )
Astrocytes   ( mesh )
Axotomy   ( mesh )
Cell Proliferation   ( mesh )
Microglia   ( mesh )
Research   ( mesh )
Telomerase   ( mesh )
Tocopherols   ( mesh )
Transduction, Genetic   ( mesh )
Dissertations, Academic -- UF -- College of Medicine -- Department of Neuroscience
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
theses   ( marcgt )

Notes

Abstract: Normal somatic cells have a finite replicative capacity, and with each cell division, telomeres (the physical ends of chromosomes) progressively shorten until they reach a critical length, at which point the cells enter cellular senescence. Some cells maintain their telomeres by the action of the telomerase enzyme. Microglia, a non-neuronal cell type residing within the central nervous system (CNS), play vital roles in maintaining neuronal function, health, and survival in both the normal and pathological CNS. Microglia are the only adult cell type in the CNS that exhibit significant mitotic potential, suggesting that these cells have limited life-spans, may rely on proliferation to replace senescent cells, and are thus susceptible to telomere shortening and subsequent cellular senescence. In our studies, we have found that telomere shortening occurred in cultured rat microglia concomitant with their progression to senescence by 32 days in vitro. Telomere shortening also occurred in vivo in both rat cerebellum and cortex from day 21 to approximately 5 months of age (i.e., the oldest age analyzed). Axotomy-activated microglia from the facial nucleus (FN) maintained telomere length (TL) via increased levels of telomerase activity (TA) during periods of high proliferation in vivo. Microglia isolated directly from the axotomized FN via fluorescence-activated cell sorting exhibited increased TA relative to un-operated controls, suggesting that microglia are the primary cell type responsible for the increased TA observed in whole tissue FN samples. Vitamin E induced a significantly high proliferation rate in cultured rat microglia. This high rate of proliferation resulted in a concomitant decrease in TL, TA, and microglial activation. Microglia retrovirally-transduced with telomerase exhibited an increased maximal life-span (ranging from 230 to 375%), and delayed entry into senescence, relative to controls and empty-vector transduced microglia. Telomerase transduction did not immortalize microglia, although these cells exhibited a normal phenotype, and had telomerase activity/protein present well past the time when all control cells had died. Our findings provide an impetus to further investigate rat microglial telomere dynamics in vivo, especially with age, following axotomy, or vitamin E supplementation, as well as in human microglia with age and incidence of Alzheimer s disease.
Subject: aging, astrocyte, axotomy, brain, life, microglia, proliferation, rat, senescence, telomerase, telomere, tocopherol, transduction
General Note: Title from title page of source document.
General Note: Document formatted into pages; contains 161 pages.
General Note: Includes vita.
Thesis: Thesis (Ph.D.)--University of Florida, 2005.
Bibliography: Includes bibliographical references.
General Note: Text (Electronic thesis) in PDF format.

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: aleph - 003322360
System ID: UFE0009641:00001


This item has the following downloads:


Full Text












ANALYSIS OF RAT MICROGLIAL CELLULAR SENESCENCE
AS DETERMINED BY MEASUREMENTS OF
TELOMERE LENGTH AND TELOMERASE ACTIVITY














By

BARRY ERIC FLANARY


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2005

































Copyright 2005

by

Barry Eric Flanary


































I dedicate this research to my beautiful wife, Allison.















ACKNOWLEDGMENTS

I thank my mentor, Dr. Wolfgang Streit, who welcomed me as a researcher into his

laboratory, permitted me to pursue the area of research most compelling to me, and who

taught me an enormous wealth of knowledge. I also thank Dr. Gerry Shaw, Dr. Satya

Narayan, and Dr. Jeff Harrison for serving on my dissertation committee and for their

scientific advice. Appreciation is also extended towards my fellow lab colleagues,

including: Chris, Amanda, Josh, Parker, Tanya, Nicole, Austin, Jackie, and Robert.

I thank Dr. Michael Fossel, who has helped open numerous doors of opportunity

for me, for all of his help throughout the years, and for all that he has done for us, in

particular for inviting me to be on the Editorial Board of his journal, The Journal of Anti-

Aging Medicine, travelling to Illinois State University (when I was an M.S. graduate

student there) to give a seminar on telomeres and cell senescence, and for inviting me to

be a plenary speaker on his scientific panel at the Inaugural International Convention on

Longevity in Sydney, Australia in March of 2004.

This thesis is written in honor of the late President Ronald Reagan (who succumbed

to Alzheimer's disease while I was working on this dissertation research), with the hope

that a cure for the disease can be found sooner rather than later.

I especially thank my parents and brother for their support, and all that they have

done to make it possible for me to pursue a path of science in life.

Special gratitude is expressed to my wife, Allison, for all of her patience,

encouragement, and support while I was working on this dissertation research.














TABLE OF CONTENTS



A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF FIGURES ................................................ viii

A B S T R A C T ...................................................................................................................... x ii

CHAPTER

1 IN T R O D U C T IO N .................................................................. .. ... .... ............... 1

Structure and Function of Telom eres ..................................................................... 1...
Characteristics of C cellular Senescence.................................................... ...............2...
The Telom ere Hypothesis of Cellular Aging........................................... ...............3...
Structure and Function of Telom erase..................................................... ...............4...
Microglial Cells of the Central Nervous System....................................................5...
Microglial Structure and Function in vitro and in vivo ...................................5...
Role of Microglia in Normal Brain Aging and Dementia...............................7...
Microglial Response Following Facial Motor Nucleus Axotomy .....................9...
N europrotective Functions of M icroglia .......................................................... 11
Specific A im s and H ypothesis................................... ...................... ............... 12

2 PROGRESSIVE TELOMERE SHORTENING OCCURS IN CULTURED RAT
MICROGLIA, BUT NOT ASTROCYTES...........................................................14

In tro d u ctio n ............................................................................................................... .. 14
M materials and M ethods ................... ....................................................... ............... 14
Culturing of M icroglia and A strocytes........................................... ................ 14
Treatm ent of M icroglial C ells ........................................................ ................ 16
D eterm nation of Telom ere Length................................................ ............... 16
Determination of Individual Chromosomal Telomere Length......................... 19
D eterm nation of Telom erase A activity ........................................... ................ 21
Determination of Cell Proliferation and Viability..........................................23
R e su lts............... .... .................... ..... ........................................................... ........ 2 4
GM-CSF Stimulates Microglial Proliferation ................................................24
Telomere Shortening Occurs in Cultured Rat Microglia ...............................25
Three-Fold Variation Exists in Individual Rat Microglia Telomeres ..............30
Cyclical Telomere Shortening Occurs in Rat Astrocytes...............................32
Telomerase Activity in Cultured Rat Microglia and Astrocytes......................38
D isc u ssio n ................................................................................................................ .. 4 1









3 TELOMERES SHORTEN WITH AGE IN RAT CEREBELLUM AND
C O R T E X IN VIV O ..................................................................................................... 50

In tro d u ctio n ................................................................................................................. 5 0
M materials and M ethods .................................................. .... ................. ................ 50
Collection of Rat Cerebellum and Cortex Tissues .........................................50
D eterm nation of Telom ere Length................................................ ................ 50
Determination of Telomerase Activity ............................................................. 51
R e su lts ..... .......................... ........ ... .......... ... ....................................... ..................... 5 1
Telomeres Shorten With Age in Rat Brain in vivo.........................................51
Telomerase Activity in Rat Cerebellum and Cortex ......................................51
D isc u ssio n ............................................................................................................... ... 5 7

4 AXOTOMY INCREASES TELOMERE LENGTH, TELOMERASE ACTIVITY
AND PROTEIN IN AXOTOMY-ACTIVATED MICROGLIA............................. 61

In tro d u ctio n ................................................................................................................ 6 1
M materials and M ethods .. ..................................................................... ................ 6 1
R at F acial N erve A xotom y ............................................................... ................ 6 1
FACS-Isolation of Rat Microglia from Micro-dissected Facial Nuclei ..............62
D eterm nation of Telom ere Length................................................ ................ 63
Determination of Telomerase Activity ...........................................................63
Telom erase W western B lot A analysis ................................................ ................ 63
H istochem istry ............................................................................................. 64
Statistical A analysis of D ata ......................................................... 65
R e su lts.................... .............. ...... ................................................................. ........ 6 5
Increase in Microglia Surrounding Axotomized Facial Nuclei........................ 65
Increase in Telomere Length in Axotomized Facial Nuclei...............................66
Increase in Telomerase Activity in Axotomized Facial Nuclei........................69
Increase in Telomerase Protein Quantity in Axotomized Facial Nuclei .............72
FACS-Isolation of Microglia from Facial Nuclei ...................... ................ 74
Increase in Telomerase Activity in FACS-Isolated Microglia From
A xotom ized F acial N uclei .......................................................... ................ 74
D isc u ssio n ............................................................................................................... ... 7 8

5 ALPHA-TOCOPHEROL (VITAMIN E) INDUCES RAPID, NON-SUSTAINED
PROLIFERATION IN CULTURED RAT MICROGLIA....................................85

In tro d u ctio n ............................................................................................................... .. 8 5
M icroglial A ctivation .. .. ................ ................................................ 85
F unction of V itam in E ......................................... ........................ ................ 86
M materials an d M eth o d s ...............................................................................................87
C ulturing of M icroglia......................................... ........................ ............... 87
Treatm ent of M icroglial C ells ........................................................ ................ 87
D eterm nation of C ell Proliferation................................................ ................ 88
Determination of Interleukin- 13 Production................................................. 88
D eterm nation of Telom ere Length................................................ ................ 88









D eterm nation of Telom erase A activity ........................................... .................. 88
Statistical A analysis of D ata ......................................................... 88
R e su lts ..................................................... ..................................................... ........ 8 9
Microscopic Examination of Cultured Rat Microglia at Various Times and
Treatm ents............................ .. ......... ................... ... .................. 89
Vitamin E Induces Cell Proliferation in Cultured Rat Microglia..................... 89
Telomere Length Analysis in Vitamin E-Treated Cultured Rat Microglia.........94
Telomerase Activity Analysis in Vitamin E-Treated Cultured Rat Microglia ....95
Interleukin-1 Beta Production in Cultured Rat Microglia...............................98
D iscu ssio n .......................................................................................................... 10 0

6 LIFE-SPAN EXTENSION IN NORMAL RAT MICROGLIA VIA
TELOMERASE REVERSE TRANSCRIPTASE RETRO VIRAL
T R A N SD U C T IO N ................................................................................................... 107

In tro d u ctio n .............................................................................................................. 10 7
M materials and M ethods ................... .............................................................. 107
C ulturing of M icroglia..................................................... ............. ............... 107
Production of Replication-Defective Telomerase-Encoding Retroviruses .......107
Transduction of Rat Microglia With Telomerase-Encoding Retroviruses........ 109
Determ nation of Telom erase Activity ...... .... ...................................... 110
Telomerase Western Blot Analysis .................................1...10
Statistical A analysis of D ata ....... ........... ............ ..................... 110
R esults............................................ ....... ... .............. .................. 110
Telomerase-Encoding Retroviral Vector................................................... 110
Telomerase Transduction Extends Life-Span of Microglia .............................11
Telomerase Activity in Transduced Microglia............................................ 113
Telomerase Protein Quantity in Telomerase-Transduced Microglia ..............117
D isc u ssio n ............................................................................................................. .. 1 1 9

7 CONCLUSIONS AND IMPLICATIONS ................. ......................................125

C o n c lu sio n s ............................................................................................................... 12 5
Im p lic atio n s .............................................................................................................. 12 8

LIST O F R EFEREN CE S .. .................................................................... ............... 130

BIOGRAPH ICAL SKETCH .................. .............................................................. 146















LIST OF FIGURES


Figure page

2-1. Cell proliferation in GM-CSF-treated cultured rat microglia as determined by
M T T an aly sis............................................................................................................ 2 5

2-2. Southern blot analysis for measurement of telomere length in cultured microglia...27

2-3. Telomere length distribution in control and GM-CSF-stimulated microglia on
d ay s 1, 16 an d 3 2 ..................................................................................................... 2 8

2-4. Southern blot analysis for measurement of telomere length in microglia cultured
at varying densities .... ..................................................................... ... .......... 29

2-5. Telomere length distribution in microglia grown to near-confluence in various
culture areas (9.5 cm 2, 21 cm 2, 175 cm 2) .......................................... ................ 30

2-6. Telomere FISH analysis of metaphase spreads of cultured rat microglia using a
FITC-conjugated peptide nucleic acid telomere-specific probe...............................31

2-7. Telomere fluorescence intensity (TFI) of all 168 individual telomeres in the 42
chromosomes of 2-day old cultures of rat microglia...........................................32

2-8. Southern blot analysis for measurement of telomere length in non-passaged
astrocytes from day 1 to 10 ....................................... ....................... ................ 33

2-9. Telomere length distribution in non-passaged astrocytes from day 1 to 10..............34

2-10. Southern blot analysis for measurement of telomere length in non-passaged rat
astrocytes from day 2 to 32 ....................................... ....................... ................ 35

2-11. Telomere length distribution in non-passaged rat astrocytes from day 2 to 32.......36

2-12. Southern blot analysis for measurement of telomere length in astrocytes from
passage 1 to 5 .............. ....................................................................... 37

2-13. Telomere length distribution in astrocytes from passages 1 to 5 .........................38

2-14. Telomerase activity in control (Con) and GM-CSF (CSF)-stimulated rat
m icroglia on the indicated days........................................................... ................ 39

2-15. Telomerase activity in non-passaged rat astrocytes on the indicated days ...........40









2-16. Quantitation of telomerase activity (arbitrary units) in rat microglia and non-
passaged astrocytes on the indicated days........................................... ................ 41

3-1. Southern blot analysis for measurement of telomere restriction fragment (TRF)
length in rat brain tissue .............. ................ .............................................. 52

3-2. TRF length distribution in rat cerebellum and cortex samples on days 21 and 152..53

3-3. Average TRF length in rat cerebellum and cortex tissues on days 21 and 152.........53

3-4. TRAP analysis for telomerase activity in rat brain tissue (days 21 to 182) ..............54

3-5. Quantitation of telomerase activity (arbitrary units) in rat cerebellum and cortex
tissues (days 21 to 182) .............. .. .............. .............................................. 55

3-6. TRAP analysis for telomerase activity in rat brain tissue (days 21 to 35) .............55

3-7. Quantitation of telomerase activity (arbitrary units) in rat cerebellum and cortex
tissues (days 2 1 to 35) .............. .............. ................................................. 56

3-8. Overall, the cerebellum exhibits higher telomerase activity than the cortex from
d ay 2 1 to 3 5 .............................................................................................................. 5 6

4-1. Micrographs of axotomized (A) and control (B) facial nucleus on day 3 post-
axotomy stained with GSI-B4 lectin to identify microglia .................................67

4-2. Southern blot analysis for measurement of telomere length in facial nuclei ..........68

4-3. Densitometric quantitation of telomere length in facial nuclei ...............................69

4-4. Representative TRAP analysis image used for measurement of telomerase
activity in facial nuclei. ............. ................ .............................................. 70

4-5. Densitometric quantitation of telomerase activity in facial nuclei.........................71

4-6. Densitometric quantitation of telomerase activity in unoperated facial nuclei .........72

4-7. Western blot image used for measurement of telomerase protein quantity in facial
n u clei ...................................................................................................... ....... .. 7 3

4-8. Densitometric quantitation of telomerase protein in facial nuclei..........................73

4-9. FACS-isolation of microglia from axotomized and control facial nuclei ..............76

4-10. TRAP analysis image used for measurement of telomerase activity in FACS-
isolated facial nuclei .... ................................................................... ........... 77

4-11. Densitometric quantitation of telomerase activity in FACS-isolated facial
n u c le i ............................................................................................................... 7 7









5-1. Representative micrographs of cultured rat microglia under various treatment
c o n d itio n s ................................................................................................................ 9 0

5-2. Cell proliferation (as determined by MTT assay) of cultured rat microglia on the
indicated days under various treatment conditions ............................. ................ 91

5-3. Proliferation rate (as determined by BrdU incorporation over 2 hours) of cultured
rat microglia on the indicated days under various treatment conditions...............93

5-4. Proliferation of cultured rat microglia at 48 hours under various treatment
c o n d itio n s ................................................................................................................ 9 4

5-5. Densitometric quantitation of telomere length in cultured rat microglia on the
indicated days under various treatment conditions ............................. ................ 95

5-6. Representative TRAP analysis image used for measurement of telomerase
activity in cultured rat m icroglia ......................................................... ................ 97

5-7. Quantitation of telomerase activity in cultured microglia on the indicated days
under various treatm ent conditions ................................................... 98

5-8. Quantitation of telomerase activity in cultured microglia at 48 hours under
various treatm ent conditions ...................................... ...................... ................ 99

5-9: Quantitation of interleukin-1 3 production by cultured rat microglia on the
indicated days under various treatment conditions ............................. ................ 99

6-1. The Clontech retroviral vector, pLPC-hTRT, used to transduce cultured rat
microglia with the human telomerase reverse transcriptase (i.e., hTRT) gene...... 112

6-2. Brightfield and green fluorescence micrographs of rat microglia and rat
glioblastoma cells (RG-2) following retroviral transduction on day 4 ................113

6-3. Representative micrographs of cultured rat microglia following retroviral
transduction on days 1 and 20 ....... ........... ............ ..................... 114

6-4. Representative photographs of hTRT-transduced cultured rat microglia on days
57 and 75 .............. ..................................................... .................................... 115

6-5. TRAP analysis image used for measurement of telomerase activity of cultured rat
microglia on the indicated days under various transduction conditions ..............116

6-6. Quantitation of telomerase activity in cultured rat microglia on the indicated days
under various transduction conditions...... .... ....................... 117

6-7. Western blot image used for measurement of telomerase protein in cultured rat
microglia on the indicated days under various transduction conditions ..............118









6-8. Quantitation of telomerase protein quantity in cultured rat microglia on the
indicated days under various transduction conditions................. .................. 118















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

ANALYSIS OF RAT MICROGLIAL CELLULAR SENESCENCE AS DETERMINED
BY MEASUREMENTS OF TELOMERE LENGTH AND TELOMERASE ACTIVITY

By

Barry E. Flanary

May 2005

Chair: Wolfgang J. Streit
Major Department: Neuroscience

Normal somatic cells have a finite replicative capacity, and with each cell division,

telomeres (the physical ends of chromosomes) progressively shorten until they reach a

critical length, at which point the cells enter cellular senescence. Some cells maintain

their telomeres by the action of the telomerase enzyme. Microglia, a non-neuronal cell

type residing within the central nervous system (CNS), play vital roles in maintaining

neuronal function, health, and survival in both the normal and pathological CNS.

Microglia are the only adult cell type in the CNS that exhibit significant mitotic potential,

suggesting that these cells have limited life-spans, may rely on proliferation to replace

senescent cells, and are thus susceptible to telomere shortening and subsequent cellular

senescence.

In our studies, we have found that telomere shortening occurred in cultured rat

microglia concomitant with their progression to senescence by 32 days in vitro.

Telomere shortening also occurred in vivo in both rat cerebellum and cortex from day 21









to approximately 5 months of age (i.e., the oldest age analyzed). Axotomy-activated

microglia from the facial nucleus (FN) maintained telomere length (TL) via increased

levels of telomerase activity (TA) during periods of high proliferation in vivo. Microglia

isolated directly from the axotomized FN via fluorescence-activated cell sorting exhibited

increased TA relative to un-operated controls, suggesting that microglia are the primary

cell type responsible for the increased TA observed in whole tissue FN samples. Vitamin

E induced a significantly high proliferation rate in cultured rat microglia. This high rate

of proliferation resulted in a concomitant decrease in TL, TA, and microglial activation.

Microglia retrovirally-transduced with telomerase exhibited an increased maximal life-

span (ranging from 230 to 375%), and delayed entry into senescence, relative to controls

and empty-vector transduced microglia. Telomerase transduction did not immortalize

microglia, although these cells exhibited a normal phenotype, and had telomerase

activity/protein present well past the time when all control cells had died.

Our findings provide an impetus to further investigate rat microglial telomere

dynamics in vivo, especially with age, following axotomy, or vitamin E supplementation,

as well as in human microglia with age and incidence of Alzheimer's disease.














CHAPTER 1
INTRODUCTION

Structure and Function of Telomeres

Telomeres are specialized structures at the physical ends of eukaryotic

chromosomes consisting of essential proteins (e.g., TRF1, TRF2, TIN2, tankyrase) and

highly conserved repeated DNA sequences (Kipling, 1995; Shay, 1999). Telomeres

control genes near chromosome ends (Wright and Shay, 1992) and may direct

chromosome attachment to the nuclear membrane (Gottschling and Cech, 1984). The

very ends of telomeres, which contain about 10 to 20 nucleotides of single-stranded

DNA, form telomere loops (T-loops) by means of a single-stranded DNA invasion event,

and are thought to protect chromosome ends from degradation and end-to-end fusions

(Shay, 1999). Vertebrate telomeres comprise the same sequence of hexanucleotide

repeats (TTAGGG)n (Moyzis et al., 1988). The length of telomeres is species-specific

and ranges from 5 to 20 kilobases (kb) in humans (Harley et al., 1990) and from 20 to

150 kb in mice (Kipling and Cooke, 1990). Telomere loss occurs with each round of

DNA replication (Harley et al., 1990) due to the inability of DNA polymerases to

completely replicate linear DNA molecules (Olovnikov, 1971, 1973, 1996; Watson,

1972), and may also occur as a result of oxidative stress (von Zglinicki, 2002). Telomere

length can be used as a predictor of the future replicative capacity of cells (Allsopp et al.,

1992), and depends on both the age of the cell and the number of times the cell has

already divided (Harley et al., 1990).









Characteristics of Cellular Senescence

Normal somatic cells undergo only a finite number of cell divisions in vitro before

entering a non-dividing state called cellular senescence (Hayflick, 1961). Senescence can

also occur in replication-independent manners, such as activation of p53 pathways (Shay

et al., 1991) or when sufficient cellular damage accumulates. Senescence, which

ultimately culminates in cell death, is characterized by an irreversible arrest of cell

proliferation (Hayflick, 1965), substantial alterations in patterns of gene expression (i.e.,

SAGE: Senescence-Associated Gene Expression) (Bernd et al., 1982; Shelton et al.,

1999; Funk et al., 2000), an increasing resistance to apoptosis (Spaulding et al., 1999),

cell-type specific changes in cell function and gene expression (Funk et al., 2000), and

concomitant telomere shortening (Harley et al., 1990). The reduction in proliferative

capacity of cells from old donors (Bowman and Daniel, 1975) and patients with

premature aging syndromes (e.g., Werner syndrome, and Hutchinson-Gilford progeria

syndrome) (Prokofeva et al., 1982), as well as the accumulation of senescent cells (Dimri

et al., 1995) both in vitro and in vivo with altered patterns of gene expression (Shelton et

al., 1999; Funk et al., 2000), implicates cellular senescence in aging and age-related

pathologies (Fossel, 2000). After cells have exhausted their replicative capacity, they

reach their Hayflick Limit and become incapable of further division (Hayflick, 1965). In

some cells, a progression of intracellular events can lead to crisis (Wright et al., 1989),

which is characterized by the appearance of one or more critically-short telomeres

(Allsopp and Harley, 1995), which activate DNA-damaging signals (Harley, 1991), and

cause end-to-end chromosomal fusions to occur (Cui et al., 2002). Thus, at least in

vertebrates, it seems that the shortest telomere length, not the average, is responsible for

maintaining chromosome stability, cell viability, and determining when a cell will enter









senescence (Hemann et al., 2001). At crisis, nearly all cells enter senescence and

ultimately die by apoptosis (Payne et al., 1994), although some are able to up-regulate

expression of the telomerase enzyme and become tumorigenic (de Lange, 1994).

Following the senescence of a cell, replication of neighboring mitotic cells can occur, and

their division to fill in the gaps left by senesced cells may cause their own telomeres to

shorten in the process. Thus, senescence can lead to a propagating cycle of accelerated

aging among remaining cells (Fossel, 2000).

The Telomere Hypothesis of Cellular Aging

The telomere hypothesis of cellular aging proposes that telomere shortening in

mitotic somatic cells contributes to and causes their senescence, hastens the senescence

of neighboring mitotic and post-mitotic cells (Harley et al., 1992), and underlies

organismal aging (Fossel, 2000). This hypothesis (Harley et al., 1992) suggests that if

telomeres in somatic cells can be maintained at/above, or increased to, pre-senescent

levels (e.g., via telomerase) in order to prevent/reverse senescence, then replicative life-

span should increase as well (Wright et al., 1996a; Fossel, 1998). If the life-span of

individual cells can be increased, then as a result, the life-span of the entire organism may

also be increased (Harley et al., 1992). Thus, if cell senescence can be slowed/prevented,

then age-related diseases may also be slowed/prevented (Fossel, 1998, 2000). The

presence of senescent cells may interfere with the normal functioning of, and may

contribute to, organ and tissue aging (Dimri et al., 1995). Telomere shortening can be

used as both an in vitro (Harley et al., 1990; Allsopp et al., 1992; Harley et al., 1992;

Flanary and Streit, 2004) and in vivo (Lindsey et al., 1991; Kajstura et al., 2000; Wright

and Shay, 2002; Flanary and Streit, 2003) marker of cell replication and cell aging.

Telomere shortening can cause changes in expression of genes nearest the telomere (i.e.,









TPE: Telomere Position Effect) (Wright and Shay, 1992; Wood and Sinclair, 2002).

Thus, as telomeres shorten with age, genes (especially those nearest the telomere) can get

over-expressed. Telomere position effect can result in the age-related expression and/or

over-expression of genes near a telomere that is dependent on both distance from the

telomere and individual chromosomal telomere length. It provides a mechanism for the

modification of gene expression that occurs throughout the replicative life span of cells

(Baur et al., 2001). The existence of TPE suggests that progressive loss of telomeres may

lead to SAGE (Dimri et al., 1995; Fossel, 1998), which may affect both cell and organ

function. Interestingly, some examples of human genes located nearest the telomere

encode for well-known age-related diseases: cataracts, neuroblastoma, prostate cancer,

Alzheimer's disease, melanoma, obesity, colorectal cancer, ovarian cancer, diabetes,

renal cell carcinoma, deafness, retinal degeneration, Huntington disease, leukemia,

coronary artery disease, breast cancer, osteoporosis, glaucoma, deafness, emphysema.

Structure and Function of Telomerase

Elongation of telomeres can occur by the action of the ribonucleoprotein enzyme

telomerase, which adds tandem hexanucleotide (TTAGGG)n repeats de novo to 3' ends of

mammalian telomeres using its own RNA as a template (Greider and Blackburn, 1985;

Morin, 1989; Cech et al., 1997). Telomerase comprises two components: an RNA

portion, which can be expressed in normal cells and is up-regulated during malignant

transformation (Blasco et al., 1996), and a protein/catalytic portion, which is a reverse

transcriptase expressed in rodent (Burger et al., 1997) and gametic/embryonic (Wright et

al., 1996b) cell types, as well as during malignant transformation (de Lange, 1994).

Telomerase can compensate for the continual shortening of telomeres that would

otherwise occur in its absence. Elongation of telomeres can result in the extension of









cellular life-span (Wright et al., 1996a; Bodnar et al., 1998). Many diverse types of

normal human and animal cell types have been transduced and subsequently

immortalized with telomerase, such as bovine adrenocortical cells (Thomas et al., 2000),

endothelial cells (Yang et al., 1999), epithelial cells (Bodnar et al., 1998), fibroblasts

(Bodnar et al., 1998), keratinocytes (Guo et al., 1998), lymphocytes (Hooijberg et al.,

2000), myoblasts (Seigneurin-Venin et al., 2000), osteoblasts (Yudoh et al., 2001), and

pancreatic islet cells (Halvorsen et al., 1999). Reconstitution of telomerase (e.g., via

retroviral transduction) in vitro into several diverse human and animal cell types can

result in restoration of replicative potential, extension of telomere length and cellular life

span, avoidance of cellular senescence (Bodnar et al., 1998; Vaziri and Benchimol,

1998), and reversion of gene expression to youthful levels (Funk et al., 2000) in the

absence of tumorigenic changes (Belair et al., 1997; Jiang et al., 1999; Morales et al.,

1999; Harley, 2002). During central nervous system (CNS) development, telomerase is

highly expressed in neural progenitor cells, but sharply decreases as synapses form, and

when cells undergo apoptosis or differentiate (Kruk et al., 1996; Mattson and Klapper,

2001).

Microglial Cells of the Central Nervous System

Microglial Structure and Function in vitro and in vivo

The central nervous system (CNS), which contains the brain and spinal cord,

contain two main populations of cells: neurons and glia. Neurons are specialized cells

important for relaying electrical signals to and from the brain and spinal cord. However,

the majority of cells present within the CNS are not neurons, but glia. Glia (i.e.,

astrocytes, oligodendrocytes, microglia) provide structural, metabolic, and trophic

support to neurons at all times. Microglia are distributed ubiquitously throughout the









central nervous system (CNS), and function as resident macrophages and antigen-

presenting cells of the CNS (Thomas, 1992). They have vital roles in supporting and

maintaining neuronal function, health, homeostasis, and survival in both the normal and

pathological CNS microenvironment (Streit, 2002a, b) by phagocytosing amyloid P3

peptide (Frautschy et al., 1992), and secreting cytokines and neurotrophic factors (Streit

et al., 1999; Nakajima et al., 2001; Streit, 2002a, b). Microglia have been aptly called

"the brain's immune system" because these cells share functional characteristics of cells

in the peripheral immune system (e.g., lymphocytes and macrophages) (Streit and

Kincaid-Colton, 1995). In addition to originating from bone marrow-derived

hematopoietic progenitor cells (Eglitis and Mezey, 1997; Hess et al., 2004), microglia are

capable of expressing MHC antigens, B- and T-cell lymphocyte markers, and other

immune cell-specific antigens.

Unlike astrocytes and oligodendrocytes, microglia are capable of significant

division, especially following neuronal injury (Kreutzberg, 1966; Graeber et al., 1988,

Svensson et al., 1994). Following acute CNS injury, there is rapid activation of microglia

and astrocytes. While acute microglial activation is marked by a conspicuous mitotic

response, reactive astrocytes undergo primarily hypertrophy with markedly enhanced

GFAP immunoreactivity, but show little mitosis (Graeber et al., 1988; Graeber and

Kreutzberg, 1986; Kreutzberg, 1996). Thus, the mitotic ability of microglia in vivo is

much greater than that of astrocytes. Interestingly, when these glial cell populations are

maintained in vitro, the mitotic potential of astrocytes exceeds that of microglia, and

astrocytes spontaneously form confluent monolayers that resemble those formed by

cultured fibroblasts. Microglia, on the other hand, require stimulation with hematopoietic









growth factors to undergo significant cell division in vitro (Giulian and Ingeman, 1988;

Suzumura et al., 1990). However, the mitotic potential of microglia both in vitro and in

vivo suggests that these cells may rely on proliferation and self-renewal to replace

senescent cells, and thus may have limited cellular life-spans.

Role of Microglia in Normal Brain Aging and Dementia

The role of microglial cells in the aging CNS and in the development of age-related

neurodegenerative disease remains unknown. Alzheimer's disease (AD) is an age-

related, progressive neurological disorder characterized by significant memory loss,

extracellular amyloid plaque deposition, intracellular neurofibrillary tangle formation

within neurons, loss of neuronal synapses, the dysfunction and death of significant

numbers of neurons, and memory loss (Mann and Yates, 1981). AD currently afflicts 1

in 10 individuals over age 65 and nearly half of those over age 85, with the incidence rate

doubling approximately every 4.4 years after age 60 (Kawas et al., 2000). Microglial

cells are known to be clustered around amyloid beta (Ap)-containing senile plaques in the

aged and AD brain (Itagaki et al., 1989), and this clustering likely occurs because the

cells are gathering there in an attempt to remove insoluble deposits of AP (Frautschy et

al., 1992). However, clearance of AP is often not achieved, and this raises the possibility

that the AP clearing ability of microglia may be weakened or lost with aging. This may

explain why substantial deposits of amyloid plaques can be found in elderly non-

demented individuals (Dickson et al., 1992). In addition, there may be overproduction of

AP such that microglia are overwhelmed by a larger-than-normal amyloid burden, which

may compromise the ability of microglia to clear amyloid, and impair their other vital

neuroprotective functions (Streit, 2002b).









Studies conducted in post-mortem human brains have shown an increased

incidence of microglial cytoplasmic structural abnormalities (i.e., cytoplasmic swelling,

twisted and shortened processes, and cytoplasmic fragmentation) and dystrophy in the

cerebral cortex of aged and AD-diseased brains (Streit et al., 2004b), which support the

hypothesis that microglia may become dysfunctional with age and that microglial

dystrophy may contribute to their senescence, which in turn, may impair their neuron-

sustaining functions and ultimately lead to neuronal cell death. It is reasonable that the

increased presence of dystrophic microglia in elderly individuals occurs because the

cells' ability to divide is declining as a result of aging (replicative senescence) thereby

slowing the replacement of senescent (dystrophic) microglia with younger cells. Based

on these observations, we hypothesize that a reduced ability of microglia to clear amyloid

with age and incidence of AD may be the result of their cellular senescence.

Understanding AD is complex and multi-faceted. Synapse loss is a hallmark

characteristic of declining memory function with aging and may be linked to an

impairment of neuronal and/or glial cell function. Neuronal integrity and function, in

turn, are highly dependent on the presence of fully functional glial cells. In the normal

CNS, microglia are engaged in the continuous monitoring of neuronal well-being (Streit,

2002b). To ensure proper neuronal functioning, complex molecular and cellular

interactions occur between neurons and microglia. Since microglia are capable of

producing both neuroprotective and neurotoxic molecules depending on the type of

signals received from neurons (Streit et al., 1999), any impairment in microglial function

due to cellular senescence (or otherwise) could have profound consequences for neuronal

activity and cognitive function in the normal aging brain. Over time, microglia may enter









senescence and be less able, or unable, to maintain neuronal health. As a result, when

sufficient quantities of microglia have begun to senesce, the neurons they once supported

may begin to degenerate, enter senescence, and ultimately die as well due to diminished

glial support and maintenance. Neuronal cell death leads to loss of communication and

synapses between neighboring neurons, and ultimately is the cause of memory loss

evident with age and in AD. Thus, neurodegenerative changes may occur because

microglia are becoming senescent and dysfunctional, and as a result, may inadvertently

contribute to neurodegeneration due to impaired glial support.

Neuronal cell death is a hallmark characteristic of AD and may be linked to an

impairment of microglial cell function. Thus, understanding how microglia are involved

in age-related deterioration of neuronal function is important for enabling the prevention

of AD. A demonstration of microglial senescence with age would suggest that slow and

progressive neurodegeneration and associated neuronal cell death, which are ultimately

responsible for memory loss and dementia, may result from diminished or impaired

microglial cell function. This could lead to the development of new drugs designed to

enhance microglial cell function and/or to slow microglial telomere shortening and

senescence as potential treatments of AD for humans.

Microglial Response Following Facial Motor Nucleus Axotomy

Even though the CNS is generally considered a post-mitotic tissue, it is important

to note that microglia do retain a robust proliferative potential, especially under

conditions of CNS injury (e.g., axotomy), as shown by DNA labelling studies using 3H-

thymidine or bromodeoxyuridine (BrdU) (Graeber et al., 1988; Kreutzberg, 1966, 1996;

Streit and Kreutzberg, 1988; Svensson et al., 1994). During facial nerve axotomy, the

facial nerve is cut outside the brain and the reactions of facial motor neurons and their









glial environment can be studied in the brainstem (Graeber et al., 1988; Kreutzberg,

1996). In the adult rat, unilateral axotomy of the facial nerve produces a robust, well-

characterized microglial response within the ipsilateral facial motor nucleus. Since the

contralateral facial nucleus is surgically unaffected, it serves as an internal control. In

adult rats, by approximately 4 weeks post-axotomy, the motor neurons of the facial

nucleus regenerate their functional connectivity (Kreutzberg, 1996), as noted by regained

whisker movement (Streit, 1996). In addition, axotomy of the facial nerve does not

disturb the blood-brain barrier (Raivich et al., 1998). After transaction of the facial

nerve, microglia but not astrocytes proliferate (Graeber et al., 1988), become

hypertrophic, and express several cell surface molecules, such as complement receptor 3,

major histocompatability complex (MHC) classes I and II (Streit et al., 1989), co-

stimulatory molecules (e.g., B7-1) (De Simone et al., 1995), and several cell adhesion

molecules (Moneta et al., 1993). Microglial cells are activated and increase in number in

the facial nucleus following peripheral axotomy. Microglia become motile and migrate

towards the injured motor neurons within the axotomized facial nucleus, and microglial

phagocytosis of bacteria can be observed in situ following axotomy (Schiefer et al.,

1999). Within a few days following axotomy, microglia also maintain close contact with

neurons and move along their dendrites, suggesting a possible role for microglia in

"synaptic stripping", the displacement of afferent synaptic terminals from the motoneuron

surface following axotomy (Kreutzberg, 1996; Schiefer et al., 1999). Under conditions of

facial-nerve axotomy, facial motor neurons survive and will eventually regenerate their

injured axons (Kreutzberg, 1996). Age does not affect the glial response to axotomy in

regards to expression of glial fibrillary acidic protein (GFAP), leukocyte common









antigen, type 3 complement receptor, and MHC classes I and II (Hurley and Coleman,

2003). The fact that microglia undergo proliferative bursts shortly after an acute injury

suggests that mitosis affords a mechanism to provide greater numbers of microglial cells,

and thus increased trophic support, during CNS injury and distress. However, their

ability to divide also suggests that the life span of microglia may be limited, and makes

them susceptible to replicative senescence (Flanary and Streit, 2004).

Neuroprotective Functions of Microglia

Microglia play both neuroprotective and immunocompetent roles which serve to

maintain neuronal health (Streit et. al., 1999; Streit, 2002b). Neurons are especially

fragile cells, and their well-being and proper functioning are highly dependent on the

presence of large numbers of microglia that sustain a plethora of neuron-supporting

functions. Microglia are extremely sensitive to even minor disturbances in CNS

homeostasis and rapidly become activated and proliferate vigorously following nearly all

neuropathologic conditions, such as nerve injury, stroke, and trauma (Streit et al., 1999).

Glial activation after injury is a beneficial and ostensibly necessary process, and serves

not only to restore homeostasis within the CNS microenvironment but also to assist in the

regeneration of injured neurons.

In addition to protecting the CNS from invading microorganisms, microglia are

important also for providing neuroprotection to normal and damaged neurons. Therefore,

it is important to sustain a healthy microglial population in order to help keep the CNS

functioning properly. During times of increased stress (e.g., acute neuronal injury),

microglia are especially important due to their unique ability to rapidly respond to

neuronal injury via migration, proliferation and trophic factor production. The

observation that there are as many microglia in the brain as there are neurons (Streit and









Kincaid-Colton, 1995), in conjunction with the fact that microglia represent the only type

of mature brain cells capable of undergoing mitosis and self-renewal, emphasizes the

importance of these cells for providing constant monitoring of neuronal well-being and

targeted trophic support to neurons that may encounter acute stress situations.

Following axotomy of motor axons within the facial nerve, neuronal survival and

axonal regeneration is accompanied by vigorous microglial activation and cell

proliferation. Thus, microglial activation, which begins long before axons have

regenerated and serves to assist in the regeneration of injured motor neurons, is an

integral, and potentially crucial, component of the regeneration process. Since

axotomized motor neurons do regenerate, the rapid onset of microglial activation likely

occurs because injured neurons are recruiting nearby microglia to assist them in their

struggle to survive and regenerate (Streit et al., 1999). These observations strongly

support a neuroprotective and pro-regenerative role of microglia in the injured CNS.

Specific Aims and Hypothesis

The specific aims are as follows:

1. To determine if cultured rat microglia are subject to telomere shortening and
senescence when cultured in vitro.

2. To determine if telomere shortening occurs in the rat brain with aging.

3. To determine if neuronal injury-induced microglial proliferation within the facial
nucleus resulted in telomere shortening in vivo.

4. Investigate long-term effects of vitamin E in cultured rat microglia.

5. To determine whether exogenous delivery of the telomerase gene via retroviral
transduction could prevent microglial senescence and extend life-span of cultured
rat microglia.

Collectively, these experiments have focused on studying the hypothesis that with

aging, microglia undergo telomere shortening both in vitro and in vivo, become






13


increasingly dysfunctional, and ultimately enter cellular senescence. The rationale for

this hypothesis is based on the fact that microglia undergo cell division in vivo, and are

thus susceptible to telomere shortening with age. If this situation does indeed occur in

vivo in multicellular telomerase-negative organisms (e.g., humans), it may lead to a

decline in microglial cell function with age, which in turn, would inhibit their ability to

promote neuronal well-being. Thus, age-related neuron loss may be due to loss of

microglial support.














CHAPTER 2
PROGRESSIVE TELOMERE SHORTENING OCCURS IN CULTURED RAT
MICROGLIA, BUT NOT ASTROCYTES

Introduction

To study the possibility that microglial cells in vitro are subject to replicative

senescence, we decided to investigate telomere shortening and telomerase activity in

microglia. We now present evidence to show that progressive telomere erosion occurs in

cultured rat microglia, while astrocytes exhibit a cyclical pattern of telomere shortening

and lengthening.

Materials and Methods

Culturing of Microglia and Astrocytes

Microglia were isolated from newborn Sprague-Dawley rat brains. The cerebral

cortices of neonatal rats (< 3 days) were stripped of meninges and minced with a sterile

scalpel blade in a 35 x 10 mm dish containing filter-sterilized 37C solution D (0.137 M

NaCl, 0.2 M NaH2PO4, 0.2 M KH2PO4, 5.4 mM KC1, 5 mM dextrose (glucose), 58.5 mM

sucrose, 0.25 [tg/mL Fungizone (Gibco, Carlsbad, CA), and 1 x 106 U

penicillin/streptomycin in sterile water). The tissue fragments/cell suspension were

incubated in 37C solution D containing 1.0% trypsin (Invitrogen, Carlsbad, CA) for 30

min. at 37C on a bi-directional tilting platform. An equal volume of Dulbecco's

modified Eagle's medium (DMEM) containing 10% fetal bovine serum (Gibco, Carlsbad,

CA) and 1% penicillin/streptomycin (complete medium) was added to quench the trypsin

reaction. The mixed brain cell suspension was then passed through a 130 |tm Nitex filter









(Tetko, Inc., Briarcliff Manor, NY) and centrifuged (4,000 rpm (2,900 g), 10 min). The

resulting pellet was resuspended in 10 mL of complete medium, passed through a 40 |tm

Nitex filter, and plated on poly-L-lysine (0.01 g/L) (Sigma-Aldrich, St. Louis, MO)

coated, solution D-rinsed, 175 cm2 flasks at a density of 1.5 brains per flask. The cultures

were incubated in complete medium at 370C under 5% CO2. After 4 days, the medium

was changed and incubation was continued for an additional 3 days. Microglia were

harvested from the whole brain cultures by shaking the flasks on an orbital shaker (100

rpm) for 1 hour (which detached the loosely-adherent microglia), and then collecting the

medium containing the free-floating microglia. The cells were then pelleted from the

medium by centrifugation (4,000 rpm (2,900 g), 10 min), resuspended in fresh complete

medium, and immediately plated (day 0) in cell culture dishes at the appropriate cell

concentrations as follows: 9.5 cm2 plates (1.0 x 106 cells/well), 3.8 cm2 plates (4.0 x 105

cells/well), or 0.32 cm2 plates (3.4 x 104 cells/well). The optimal initial cell plating

density was empirically determined in previous experiments. Cells were allowed to settle

for 1 hour in a 5% C02, 370C incubator, and then the culture medium was changed to

remove any contaminating non-adherent cells. The microglia were then treated with the

appropriate concentration of a particular treatment regimen.

Astrocytes normally form monolayers that cover the bottom of the culturing flask

containing mixed brain cell cultures. To prepare enriched astrocyte cultures, all cells

adhering to the astrocytic monolayer were detached by vigorously shaking the flasks at

200 rpm for 1 hour. The culture medium containing the floating cells was then removed,

and the remaining adherent astrocytes were rinsed with PBS, trypsinized, counted, and

plated. Astrocytes were plated on day 0 at an initial density of Ixl06 cells/well in 9.5









cm2 plates and allowed to divide, or at a density of 2xl06 cells in 175 cm2 flasks and

passage when confluent.

Treatment of Microglial Cells

Microglia were treated on day 0 with either 1.0 nM (0.015 [tg/mL) or 10.2 nM

(0.15 [tg/mL) recombinant rat granulocyte-macrophage colony stimulating factor (GM-

CSF) (R&D Systems, Minneapolis, MN),), 100 nM lipopolysaccharide (LPS), or

received no stimulation (control). In all experiments, media (and respective treatment)

were changed as needed (usually every 3 to 4 days).

Determination of Telomere Length

To measure telomere length (i.e., telomere restriction fragment (TRF) length: the

length of the telomere plus sub-telomeric DNA, the latter being dependent upon the

particular cleavage sites of the two restriction enzymes used), Southern blot analysis

using chemiluminescent detection and the DIG-High Prime DNA Labeling and

Detection Starter Kit II (Roche, Indianapolis, IN) was employed as previously described

(Flanary and Streit, 2003, 2004), with minor modifications. Genomic DNA was isolated

from 1) cultured cells at various time points, 2) whole brain tissue samples, or 3) pooled

micro-dissected facial nuclei (approximately 7 mg wet weight each), using the DNeasy

DNA isolation kit (Qiagen, Valencia, CA). DNA concentration was determined by

absorbance at 260 nm, while DNA purity was calculated by the ratio of 260/280 nm

absorbance, using a spectrophotometer. For Southern blotting, either 2.5 |tg (for cultured

cells), or 5.0 |tg (for tissues) of DNA was digested with 10 units each of HinjI and Rsal

overnight at 370C. Following digestion, 5.0 ptL gel loading dye (0.25% xylene cyanol,

0.25% bromophenol blue, 30% glycerol) was added to each sample. Each digested DNA









sample, and two samples containing a digoxigenin-labeled molecular weight (MW)

marker (100 ng each lane) (i.e., DIG DNA MW marker II) (Roche, Indianapolis, IN),

were then loaded onto a horizontal 15 x 25 cm 0.5% agarose gel, and electrophoresed at

70 volts in IX TBE buffer at 40C with buffer recirculation until the bromophenol blue

band ran off the gel and the xylene cyanol band reached 80% the length of the gel

(approximately 21 hours). The gel was then soaked successively in depurination solution

(0.25 M HC1) for 5 min, then denaturation solution (1.5 M NaCl, 0.5 M NaOH) for 2 x 10

min., and finally in neutralization solution (1.5 M NaCl, 0.5 M Tris) for 2 x 10 min.

Unless otherwise noted, all incubations and washes were performed at room temperature

with gentle agitation. The gel was rinsed three times in sterile double-distilled water after

each treatment noted above, and was then equilibrated in 20X SSC (3.0 M NaCl, 0.3 M

sodium citrate; pH 7.0) for 10 minutes before the DNA was vacuum blotted (Boekel,

Feasterville, PA) onto a positively-charged nylon membrane at 45 mbar for 45 min.

Following UV-crosslinking (120 mJ/cm2), a 2 min. wash in 2X SSC, and

prehybridization at 41C for 1 to 2 hours, hybridization of telomeric repeats was

accomplished by using a digoxigenin-labeled telomere-specific oligonucleotide probe

(TTAGGG)3. Digoxigenin labeling of the probe (100 pmol) was accomplished by using

the DIG oligonucleotide tailing kit (Roche, Indianapolis, IN). Following hybridization at

410C overnight (approximately 17 hours) in a hybridization oven (Hybaid, Franklin, MA)

with gentle rotation, the membrane was washed in 2X wash buffer (2X SSC, 0.1% SDS)

for 2 x 5 min., followed by washes in 0.5X wash buffer (0.5X SSC, 0.1% SDS) for 2 x 15

min. at 550C in a hybridization oven with moderate rotation. After washing the

membrane in washing solution (0.1 M maleic acid, 0.15 M NaCl, 0.3 % (v/v) Tween-20;









pH 7.5) for 2 min., the membrane was then soaked in blocking solution for 1 to 2 hours,

followed by incubation with a 1:10,000 dilution of an alkaline phosphatase (AP)-

conjugated anti-digoxigenin antibody for 30 to 45 min. Following washes in washing

solution for 2 x 15 min., chemiluminescent detection was accomplished by washing the

membrane in detection solution (0.1 M NaC1, 0.1 M Tris; pH 9.5) for 3 min., and then

incubating the membrane (wrapped in plastic wrap) in the AP-metabolizing substrate

CDP-Star (Roche, Indianapolis, IN) for 5 min. at 37C. Following exposure of the

membrane to X-omat AR Film (Eastman Kodak Company, Rochester, NY) in an

autoradiography cassette for 1 to 60 min., the film was developed using a Konica SRX-

101A automatic film processor (Konica Minolta, Mahwah, NJ). A digital image of the

autoradiograph was generated by scanning it using a GS-710 calibrated imaging

densitometer (BioRad, Hercules, CA). Telomere lengths (shortest, mean, longest) of

each sample were calculated by comparison to known MW standards present on the gel

(in each outside lane), and quantified using the computer program Telometric (version

1.2) (Grant et al., 2001).

Telomere lengths (shortest, mean, longest) of each sample were determined as

follows. The length of the "shortest" telomeres represented the telomere signal (smear)

on the autoradiograph corresponding to the smallest telomeres (lowest MW), which was

calculated by measuring the bottom of the smear in each lane. Similarly, the length of the

"longest" telomeres represented the signal corresponding to the longest telomeres

(highest MW), which was calculated by measuring the top of the smear in each lane. The

length of the "mean" telomeres represented the signal corresponding to the average

length of all telomeres within the entire length of the smear. Telomere length









measurement by current Southern blot techniques normally, and unavoidably,

incorporates sub-telomeric DNA regions into the calculated MW due to the use of DNA

digestion via restriction enzymes. Sub-telomeric DNA regions are located directly

adjacent to the telomere regions of DNA, and are still connected to the telomere region

during gel electrophoresis due to the action of restriction enzymes, which do not cleave

directly at the telomere/sub-telomeric region. During gel electrophoresis, the presence of

these small sub-telomeric DNA regions will slow the migration of telomere regions,

which would otherwise migrate slightly faster in the absence of such attached sub-

telomeric regions. Thus, all protocols used for Southern blot for telomere length analysis

utilizing restriction enzymes do not generate a "true" measurement of "actual" telomere

length, since the pure telomere DNA region (if unconnected to adjacent sub-telomeric

regions) would migrate faster on the gel, and thus would have a slightly smaller actual

MW than what was measured.

Determination of Individual Chromosomal Telomere Length

Fluorescence in situ hybridization (FISH) is a molecular cytogenic technique that is

used to obtain information from metaphase (Poon et al., 1999) or interphase (De Pauw et

al., 1998) cells, depending on the specific sequence of the fluorochrome-conjugated

probe applied. We used a telomere-specific FITC-conjugated probe, and the binding of

the probe to its target (telomeres) can be identified by a distinct green fluorescence signal

at the tips of metaphase chromosomes. For FISH analysis, metaphase chromosomes were

obtained from cultured microglia using standard methods. Briefly, fresh cultures (day 0)

of microglia were stimulated to divide with 10 nM rrGM-CSF for 2 days on top of glass

coverslips. Colchicine (10 [tg/mL) was added, and the cells were incubated for 1 hr. at









370C. This step disrupts and prevents formation of mitotic spindles, prevents completion

of mitosis, and enriches the population of metaphase cells. Following aspiration, pre-

warmed (to 370C) 0.075 M KCl was added, which makes nuclei swell osmotically and

helps prevent chromosome overlap, and the cells were incubated for 20 min. at 37 C.

The cells were then fixed with ice-cold methanol/acetic acid (3:1). A FITC-conjugated

peptide nucleic acid (PNA) telomere-specific probe (Dako, Carpinteria, CA) was added,

which was used due to its high sensitivity and specificity. PNA is a synthetic DNA/RNA

analog capable of binding 99 to 100% of telomere repeats. Additionally, this probe does

not recognize subtelomeric sequences and, therefore, will allow for an exact

measurement of telomere length. Chromosomes were counterstained with 100 mg/mL

propidium iodide and mounted with the antifade reagent Vectashield. The preparations

were viewed with both a Zeiss Axioskop 2 fluorescence microscope connected to an RT

color Spot digital camera (Diagnostic Instruments, Inc., Sterling Heights, MI) using a

60X and 100X oil lens, and a Bio-Rad 1040 ES confocal system connected to an

Olympus IX70 inverted microscope using an Olympus planapo 60X 1.40 oil lens. High

quality preparations were photographed using a digital camera at 1024 x 1024 resolution.

Using special software, telomere length was measured at the ends of individual

chromosomes from digital images of metaphase spreads (Poon et al., 1999) using the

Zeiss fluorescence microscope. Using Image-Pro Plus software (Media Cybernetics,

Carlsbad, CA), individual telomere fluorescence intensity was measured. This enabled a

determination of whether intrachromosomal telomere length variation occurs in these

cells, and if certain chromosomes are more susceptible to shortening over time.









Determination of Telomerase Activity

Telomerase activity was measured using the telomere repeat amplification protocol

(TRAP) as previously described (Flanary and Streit, 2003, 2004), with minor

modifications. Total protein was isolated from 1) PBS-washed cultured cells, 2) whole

brain tissue samples, or 3) pooled micro-dissected facial nuclei (approximately 0.5 mg

wet weight each), using 200 ptL CHAPS lysis buffer (5.0 mM P-mercaptoethanol, 1.0

mM EGTA, 1.0 mM MgCl2, 0.1 mM phenylmethylsulfonyl fluoride, 10 mM Tris, 0.5%

CHAPS, 10% glycerol). The protein extract solution was collected in RNAse-free tubes,

incubated on ice for 30 min., and centrifuged at 12,000 g (14,000 rpm) for 20 min. at 4C

to sediment residual cell debris, if present. Protein extracts were then aliquoted into

RNAse-free tubes and stored at -800C. Immediately prior to TRAP analysis, total protein

concentration was measured using the BCA protein assay reagent (Pierce, Rockford, IL)

as per the manufacturer's recommended protocol. For TRAP analysis, each sample set

included normal protein extracts, a telomerase-negative control (CHAPS lysis buffer,

and/or RNAse-treated extracts: 10 mg/mL RNAse:sample (1:1) incubated for 20 min. at

room temp.), and a telomerase-positive control (500 ng protein extract of a rat

glioblastoma cell line RG-2). Each 50 ptL reaction initially contained 5.0 pL 10X TRAP

buffer (10 mM EGTA, 500 mM KC1, 15 mM MgCl2, 100 mM Tris, 1.0 mg/mL BSA,

0.05% Tween-20), 200 [tM dNTP (Roche, Indianapolis, IN), 100 ng telomerase substrate

(TS) primer (5'-AAT-CCG-TCG-AGC-AGA-GTT-3'), 500 ng protein extract, and

RNAse-free water up to 48 [tL. This mixture was incubated for 20 min. at room

temperature to allow telomerase, if present and active, to add hexanucleotide telomeric

repeats (i.e., TTAGGG) onto the 3' end of the TS primer, which is a substrate









oligonucleotide and served as an artificial telomere. Following telomeric extension, 100

ng CX primer (5'-CCC-TTA-CCC-TTA-CCC-TTA-CCC-TAA-3') and 5 units Taq

polymerase (Fisher Scientific, Pittsburgh, PA) were added. Telomere repeats were

amplified by the polymerase chain reaction (PCR) using the TS (forward) and CX

(reverse) primers, which generate a 40 base pair internal control band (i.e., TS-CX primer

dimer) in each lane (including the negative control lane, since the presence of the internal

control band is independent of the activity of telomerase), and a ladder of products

(generated by telomerase) containing 6-base increments beginning at 46 base pairs in

telomerase-positive lanes. Both an increased quantity and intensity of bands present

within the ladder of products correspond to an increased level of telomerase activity. If

telomerase activity was absent, no product ladder was formed and only the internal

control band was evident. PCR was carried out as follows: Initial denaturation at 94C

for 2 min. to inactivate telomerase, then 33 total cycles of the following: 940C for 30

sec., 550C for 30 sec., 720C for 45 sec. Following PCR, 5 ptL filter-sterilized gel loading

dye (0.25% xylene cyanol, 0.25% bromophenol blue, 50% glycerol, 50 mM EDTA) was

added to each sample. The samples were then loaded onto a vertical 20 cm 12.5% non-

denaturing polyacrylamide gel, and electrophoresed at 87 volts at room temperature in

0.5X TBE buffer until the bromophenol blue band ran off the gel and the xylene cyanol

band reached 95% the length of the gel (approximately 21 hours). The telomerase

products were visualized by staining the gel with a 0.01% solution of SYBR Green

(Molecular Probes, Eugene, OR) for 40 min. in the dark with gentle agitation, and then

photographing the gel under ultraviolet light using an electronic gel documentation

system (Gel Doc 2000, BioRad, Hercules, CA). Quantitation of telomerase activity (i.e.,









the ladder of products formed in each lane) was performed using the densitometry

computer program Quantity One (version 4.3.1) (BioRad, Hercules, CA). Average

telomerase activity was determined by calculating the mean of individual quantitative

measurements from two or more identical samples. Normalized telomerase activity was

determined by comparing the average densitometric values of identical samples run on

different gels in order to make an accurate comparison of telomerase activity between all

samples run on multiple gels.

Determination of Cell Proliferation and Viability

To assess cell proliferation, two methods were employed: the colorimetric MTT

assay, and 5-Bromo-2'-deoxyuridine (BrdU) incorporation. For the MTT assay

(Boehringer Manheim, Indianapolis, IN), which measures both cell proliferation and cell

viability, microglia were plated at an initial density of 3.4 x 104 cells/well in 0.32 cm2

(96-well) plates. MTT labeling reagent (5 mg/mL of 3-[4,5-dimethylthiazol-2-yl]-2,5-

diphenyl tetrazolium bromide in PBS) (Boehringer Manheim, Indianapolis, IN) was

added (10 ptL per well) to the cultured cells at various time points. Metabolically-active

cells cleave the yellow tetrazolium MTT salt to form purple formazan crystals via NADH

reductase. Following a 4 hour incubation at 370C under 5% CO2, cells were solubilized

overnight in 10% SDS in 0.01 M HCI (100 ptL per well). The solubilized formazan

product was spectrophotometrically quantified at 550 nm (using a reference wavelength

of 655 nm) using a Benchmark microplate reader (BioRad, Hercules, CA) and microplate

manager software (version 4.0).

To determine proliferation using BrdU incorporation, microglia were plated at an

initial density of 2.0 x 105 cells/well in 1.9 cm2 plates (24-well plates). BrdU (Sigma-









Aldrich, St. Louis, MO; catalog # B5002) was added (10 ptM final concentration) to the

cultured cells at various time points. Proliferating cells incorporated the BrdU (in place

of thymine) during S-phase. Cells were fixed at 40C overnight in 80% EtOH. The

following morning, wells were rinsed with PBS and stored at 40C in PBS until assayed.

For BrdU analysis, each plate was incubated for 10 minutes at 370C in 2M HC1.

Following washes in PBS for 3 x 5 min., blocking buffer (PBS containing 0.1% Triton X-

100 and 2% normal goat serum) was added, and each plate was incubated for 30 min. at

370C. A FITC-conjugated rat anti-BrdU antibody (Serotec, Raleigh, NC; catalog #

MCA-2060-FT) diluted in blocking buffer was added to each well, and each plate was

incubated for 2 hours at room temperature covered with foil (to prevent degradation of

fluorescent signal). Wells were rinsed with PBS, and cell nuclei were stained with a 1.0

[tg/mL solution of 4'-6-Diamidino-2-phenylindole (DAPI) for 5 min. covered with foil.

Cells were photographed under fluorescence using a digital camera (Sony DSC-S75

Cyber-shot, 3.3 megapixels, Carl Zeiss Vario-Sonnar lens) connected to a Zeiss Axiovert

25 fluorescence inverted microscope.

Cell viability was performed on cultured microglia using the live/dead

viability/cytotoxicity kit (Molecular Probes, Eugene, OR).

Results

GM-CSF Stimulates Microglial Proliferation

Microglia exposed to GM-CSF increased proliferation compared to controls (Fig.

2-1). Cell viability analysis of cultured microglia indicated that with increasing time in

culture, the total number of viable cells decreased in both control and GM-CSF-treated

groups, indicating that the cells were entering replicative senescence (data not shown).









0.60
0.55
0.50
t0 0.45
0.40
0 0.35
B 0.30
O* 0.25
0.20
S0.20
0.15
S0.10
( 0.05
0.00


-U- CSF (10 n \l)
Control















0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42
Day


Figure 2-1. Cell proliferation in GM-CSF-treated cultured rat microglia as determined by
MTT analysis. Following stimulation with the mitogen rrGM-CSF (CSF),
microglia undergo a significant burst (p<0.001) in proliferation from day 2 to
9. However, when cultured in the continual presence of CSF, microglia
undergo rapid telomere shortening (Flanary and Streit, 2004), and exhibit a
significant decrease (p<0.001) in proliferation (when comparing day 9 to 41,
day 16 to 41, or day 27 to 41). Importantly, proliferation in CSF-stimulated
microglia eventually falls to levels below that evident in un-stimulated control
microglia (i.e., on day 41). This suggests that microglia in vitro undergoing
continual rapid division apparently "use up" most of their replicative potential
within the first 14 days (as evidenced by the smaller burst in proliferation
from dayl6 to 27 (p>0.05) compared to day 2 to 9), and are no longer able to
sustain this rate at later timepoints, at which point they likely succumb to
replicative senescence.

Telomere Shortening Occurs in Cultured Rat Microglia

Both control and rrGM-CSF (GM-CSF)-stimulated microglia underwent telomere

shortening while in culture (Fig. 2-2). In control microglia, both the longest and mean

telomere length remained relatively unchanged from day 1 to 16 while the shortest

telomeres decreased a moderate 6.3 kb. In contrast, GM-CSF-stimulated microglia

exhibited dramatic telomere loss (10.8 kb) in the longest telomeres from day 1 to 16, with









minimal shortening (0.9 kb) occurring from day 16 to 32. Mean telomere length

decreased only slightly over time. Conversely, the quantity of short telomeres greatly

increased (9.1 kb) from day 1 to 16, with a small increase (2.1 kb) occurring from day 16

to 32 (Fig. 2-3). Thus, in stimulated microglia, the longest telomeres were allowed to

shorten while short telomeres were lengthened. Additionally, the shortest telomeres are

continually being lengthened over time, even immediately prior to senescence (around

day 32). The quantity of longest telomeres was much larger (9 kb) in GM-CSF-treated

microglia than in controls on day 1 (Fig. 2-3). This indicates that rapid telomere

lengthening occurs in microglia during the first 24 hours of being exposed to GM-CSF,

and is supported by TRAP data which shows that telomerase activity is substantially

higher in stimulated microglia, compared to unstimulated microglia. No DNA was

available to be isolated on day 32 in control cells, and very little in GM-CSF-treated

cells, since they were already senescent at this time point. When microglia were plated in

various sized culture dishes and grown to confluence, telomere length diminished

proportional to the size of the culture dish. With increasing culturing area (i.e., 9.5 cm2,

21 cm2, and 175 cm2), telomere length decreased in the longest (1.4 kb), mean (1.3 kb),

and shortest (2.1 kb) telomeres (Fig. 2-4, 2-5).







(0
rW


WO C
qrr i- M


C C LL LL LL
0 0 00) CO)
00000
U^^ U^^ U ^ (a) (a


TI


23.1 kb--


9.4 kbf-


6.6 kb--m


4.4 kb-- -
Figure 2-2. Southern blot analysis for measurement of telomere length in cultured
microglia. Genomic DNA was isolated on the indicated days (1, 16, 32) in
cells treated with either sterile water (CON), or 10 nM rrGM-CSF (CSF). For
both treatments, telomere length decreased over time. However, telomere
shortening was much more pronounced in GM-CSF-treated cells.


aM










CON 1


CON 16


CSF 1


I


CSF 16


CSF 32


El Longest 32.9 33.4 41.9 31.1 30.2
* Mean 29.6 29.4 32.7 30.2 29.0


m Shortest 22.3 16.0 14.4
Figure 2-3. Telomere length distribution in control and
days 1, 16, and 32.


23.5 26.6
GM-CSF-stimulated microglia on


I


I


I


-ram


- m m









0) N)

uM P


- 23.1 kb


9.4 kb


6.6 kb
Figure 2-4. Southern blot analysis for measurement of telomere length in microglia
cultured at varying densities. Cells were plated without GM-CSF at varying
densities and culture areas as follows: 5x105 cells (9.5 cm2), 5x105 cells (21
cm2), or 2x106 cells (175 cm2). Genomic DNA was isolated once cells
reached near-confluence following 3 days of growth. Increased telomere
attrition is evident in microglia plated in larger culture areas evidently caused
by their increased division, which was necessary in order to fill the available
culturing area.
































* Shortest 25.8 26.9 23.7


Figure 2-5. Telomere length distribution in microglia grown to near-confluence in
various culture areas (9.5 cm2, 21 cm2, 175 cm2).

Three-Fold Variation Exists in Individual Rat Microglia Telomeres

Individual microglia telomeres were visualized using FISH analysis (Fig. 2-6).

Telomere fluorescence intensity (TFI) is a measurement (arbitrary units) used to provide

an indicator of the number of telomeric repeats present on a particular telomere.

Microglia on day 2 of in vitro growth were found to have a normal karyotype consisting

of 42 chromosomes. The TFI was found to vary greatly both between and within

individual chromosomes (Fig. 2-7). Variation in TFI values on all chromosomes ranged

from 17.0 to 52.0, representing a 3.1-fold maximal difference in telomere length (mean

TFI = 44.2 + 5.2). Variation of TFI values on the same chromosome ranged from 17.6 to

47.0, representing a 2.7-fold difference in telomere length.















































Figure 2-6. Telomere FISH analysis of metaphase spreads of cultured rat microglia using
a FITC-conjugated peptide nucleic acid telomere-specific probe. Cells were
viewed at 90X (left) and 270X (right) magnification via confocal microscopy.
Note the green telomeres at the ends of red chromosomes. Two large red cell
nuclei containing interphase chromosomes are visible in the upper portion of
the 90X view. Substantial telomere length variation was found to exist both
between and within individual microglial telomeres. Scale bar is
approximately 10 |tm (in 90X view) and 5 |tm (in 270X view).









80


70




50-



o 40


30


20




0

17 24 31 38 45
Telomere Ruorescence Intensity

Figure 2-7. Telomere fluorescence intensity (TFI) of all 168 individual telomeres in the
42 chromosomes of 2-day old cultures of rat microglia. TFI is an arbitrary
measurement corresponding to the number of telomeric repeats present on
each telomere. Thus, TFI is directly proportional to telomere length. Note the
large quantity of long telomeres, as evidenced by high TFI values, which are
present in young microglial cultures.

Cyclical Telomere Shortening Occurs in Rat Astrocytes

Astrocytes were also found to undergo telomere shortening during the first 10

days of in vitro culturing (Fig. 2-8). From day 1 to 3, telomere length decreased rapidly,

with the differences being 12.7 kb in the longest, 6.1 kb in the mean, and 13.8 kb in the

shortest telomeres. Beginning on day 3, a fluctuating pattern in telomere length in all

groups (i.e., longest, mean, shortest) begins to develop. From day 3 to 4, telomeres in all







groups (except the longest) are lengthened, most notably the shortest (10 kb). However,
from day 4 to 5, telomeres in all groups shorten. From day 5 to 7, telomeres in all groups
are again lengthened. Then, from day 7 to 10, telomeres in all groups (except the longest
from day 8 to 9) shorten again (Fig. 2-9).


1
U,


345678910
*-^-- wee^


23.1 kb-




9.4 kb -


lom


Figure 2-8. Southern blot analysis for measurement of telomere length in non-passaged
astrocytes from day 1 to 10. Telomere length decreased over time, especially
during days 1 to 3. A cyclical pattern of telomere shortening and re-
lengthening was observed from day 1 to 10.

































Shortest 32.9 26.2 19.1 29.1 26.6 29.4 30.6 30.4 28.2 26.9

Figure 2-9. Telomere length distribution in non-passaged astrocytes from day 1 to 10.
Note the cyclical pattern of telomere lengthening and shortening that occurs
with time in culture.


During lengthening, the shortest telomeres are lengthened the quickest, suggesting that

maintenance of the shortest telomeres is most critical (Hemann et al., 2001). Astrocytes

were also found to have consistently longer telomeres in all groups compared to

microglia. Mean telomere length in astrocytes on day 1 is 6.1 kb longer than in GM-

CSF-treated microglia, and 9.2 kb longer than in control microglia on the same day. The

longest telomeres in astrocytes on day 1 are 7.7 kb longer than in GM-CSF-treated

microglia, and 16.7 kb longer than in control microglia on the same day. Similarly, the

shortest telomeres in astrocytes on day 1 are 10.6 kb longer than in control microglia, and

18.5 kb longer than in GM-CSF-treated cells on the same day (Fig. 2-3 and 2-9). Long-








term growth of cultured astrocytes (up to day 32) revealed a similar pattern of telomere

lengthening followed by a period of telomere erosion (Fig. 2-10). From day 2 to 16,

telomere length increased in the longest (4.9 kb), mean (4.1 kb), and shortest (1.1 kb)

telomeres. Similarly, from day 16 to 32, telomeres shorten in the longest (3.5 kb), mean

(3.0 kb), and shortest (1.2 kb) telomeres (Fig. 2-11). This cyclical pattern of telomere

lengthening and shortening in astrocytes is supportive of a similar cyclical pattern of

transient telomerase expression and repression that is occurring in these cells.




2 9 16 24 32












23.1 kb-







9.4 kb- -

Figure 2-10. Southern blot analysis for measurement of telomere length in non-passaged
rat astrocytes from day 2 to 32. Telomere length increased from day 2 to 16,
then declined from day 16 to 32. This cyclical pattern of telomere lengthening
and attrition correlates to increasing and decreasing telomerase activity over
the same time period.










35-



S25--
S20--



S15--


0


0 -


l Longest 27.2 29.5 32.1 28.2 28.6
Mean 25.7 27.5 29.8 26.9 26.8
E Shortest 24.3 25.3 25.4 17.7 24.2

Figure 2-11. Telomere length distribution in non-passaged rat astrocytes from day 2 to
32. Note the cyclical pattern of telomere shortening and lengthening that
occurs with time in culture.

When astrocytes are passage serially, mean telomere length remains nearly

unchanged from passage 1 to 5 (Fig. 2-12). However, both the longest and shortest

telomeres undergo a cyclical pattern of telomere lengthening and shortening. Both sets of

telomeres lengthen from passage 1 to 2, and then shorten from passage 2 to 3.

Subsequently, they re-lengthen from passage 3 to 4, and then shorten from passage 4 to 5.

The shortest telomeres exhibit the most dramatic lengthening. From passage 1 to 2 they

lengthen 7.4 kb, and from passage 3 to 4 they lengthen 9.7 kb. The overall trend for the

shortest telomeres is a continual lengthening over time (Fig. 2-13).








12345
OR^nl^


23.1 kb -


9.4 kb- F F-


6.6 kb -
Figure 2-12. Southern blot analysis for measurement of telomere length in astrocytes
from passage 1 to 5. The interval between each passage was as follows:
passage 1 to 2 (4 days), 2 to 3 (6 days), 3 to 4 (6 days), 4 to 5 (9 days). While
the mean telomeres remain nearly unchanged, the longest and shortest
telomeres undergo a cyclical pattern of telomere lengthening and shortening,
with the shortest telomeres gradually lengthening over time.









35-

.0 3 0 ---- ~ ------------





20-


_o



0-


E Longest 29.4 31.6 29.8 31.0 28.3
Mean 27.0 27.7 27.7 27.7 27.2
E Shortest 11.8 19.2 17.6 27.3 26.7

Figure 2-13. Telomere length distribution in astrocytes from passages 1 to 5. Note the
cyclical pattern of telomere shortening and lengthening that occurs with time
in culture.

Telomerase Activity in Cultured Rat Microglia and Astrocytes

Telomerase activity was measured in both microglia and astrocytes on various

days (Fig. 2-14 to 2-16). Telomerase activity was consistent, and at a low level, in

control microglia from day 0 to 32. In contrast, GM-CSF-stimulated microglia express a

nearly 3-fold increase in telomerase activity on day 2, compared to controls.

Subsequently, telomerase activity follows a cyclical pattern and declines until day 24,

then increases dramatically again from day 24 to 32 (Fig. 2-14 and 2-16). In astrocytes,

telomerase activity gradually increased from day 0 to 2, then rapidly increased from day

2 to 16. Telomerase activity then dramatically declined from day 16 to 24. By day 32,

activity was nearly non-existent (Fig. 2-15, 2-16). Telomerase activity correlated well









with telomere length in astrocytes. Mean telomere length was found to increase in

astrocytes by 4.1 kb from day 2 to 16, corresponding to a period of high telomerase

activity, especially on days 9 to 16 (during which time telomeres were lengthened).

Subsequently, mean telomere length decreased 3.0 kb from day 16 to 32, corresponding

with a period of low/absent telomerase activity (during which telomeres shortened) (Fig.

2-11 and 2-16).


-"-M !r, N M












*a U *. U. U.






U f-J
I iC








Figure 2-14. Telomerase activity in control (Con) and GM-CSF (CSF)-stimulated rat
microglia on the indicated days. Telomerase activity in control microglia
remains relatively consistent. However, in GM-CSF-stimulated microglia,
telomerase activity is increased dramatically on days 2 and 32, compared to
controls. (-) = telomerase-negative control; (+) = telomerase-positive control;
IC = internal control.












D It C4 +-.4




I.






I j


*' a .





















IC


Figure 2-15. Telomerase activity in non-passaged rat astrocytes on the indicated days.
Telomerase activity in astrocytes exhibits a cyclical pattern of increasing and
decreasing telomerase activity. (-) = telomerase-negative control; (+) =
telomerase-positive control; IC = internal control.









90
> 80
70
60
50
40
30 -A- Astrocyte
20 -0- Microglia Con
101 -- Microglia CSF
0 -0-- Positive Con

0 10 20 30 40
Day

Figure 2-16. Quantitation of telomerase activity (arbitrary units) in rat microglia and
non-passaged astrocytes on the indicated days. Control microglia express low
telomerase activity than the positive control. GM-CSF-stimulated microglia
exhibit a cyclical cycle of increasing and decreasing telomerase activity.
Telomerase activity in astrocytes on day 0 is nearly 3-fold higher than in
microglia on the same day and was always higher than the positive control.
Additionally, GM-CSF-stimulated microglia exhibit a 3-fold higher level of
telomerase activity on day 2, compared to controls.
Discussion
We have shown that both microglia and astrocytes undergo dynamic changes in

both telomere length and telomerase activity with time in culture. While telomere

shortening occurs gradually in actively dividing microglia and is accompanied by their

progression into senescence, astrocytes exhibit a cyclical pattern of telomere lengthening

and shortening and are able to divide for much longer periods of time in vitro than

microglia. Our study, which is the first to report on telomere length and telomerase









activity in glial cells of any organism, shows that telomeres shorten progressively in

microglia with increased cell division and with time in culture.

The fact that microglia can be induced to proliferate following CNS injury suggests

that microglia represent a self-renewing population of cells. Tour current findings open

the possibility that the replicative potential of microglia in vivo is limited and that these

cells may at some point exhaust their replicative capacity. Thus, injury-induced mitosis

of microglia in vivo could result in telomere shortening, which may drive the cells down

an accelerated path towards cellular senescence.

Rapid elongation of the shortest telomeres in GM-CSF-treated cells occurs from

day 1 to 16, suggesting that these cells are trying to evade senescence by maintaining the

shortest telomeres above the critical length. The shortest telomeres lengths in GM-CSF-

treated cells are 14.4 kb on day 1, while controls have lengths of 16 kb on day 16 (and

enter senescence shortly thereafter). Thus, the critical telomere length that triggers

senescence appears to be somewhere slightly below 14.4 kb. Control microglia exhibit

telomere attrition of the shortest telomeres, while the longest and mean telomeres remain

nearly unchanged, from day 1 to 16, which suggests that either the longest telomeres are

being preferentially maintained, or that non-uniform telomere lengthening occurs such

that the longest telomeres actually increase in size (by 0.5 kb) while the shortest continue

to erode (by 6.3 kb). The data also suggests that attrition of the shortest telomeres occurs

in the presence of low levels of telomerase (since both longest and mean telomeres are

maintained). This is supported by our TRAP data indicating that telomerase is present in

low amounts, which is likely enough to maintain the longest and mean telomeres, while

the shortest telomeres continually erode. Unstimulated cells appear to enter senescence









when telomeres reach critical lengths, however, the GM-CSF-stimulated microglia do

not. No critical length is reached in these stimulated cells, yet they still undergo

senescence, as evidenced by reduced mitotic activity and viability after several weeks in

culture, suggesting that mechanisms other than critically-short telomeres (e.g., ROS

damage, cellular trauma, shortening of the longest telomeres) may be triggering

senescence in these cells in the absence of telomerase repression (Kodama et al., 2001).

Telomere erosion is also thought to occur directly by other means, which are

independent of cell replication (von Zglinicki, 2002). Since GM-CSF-treated microglia

on day 1 have longest telomeres that are 9 kb longer than in controls on the same day,

GM-CSF may up-regulate high telomerase activity during the first few days of growth

(Szyper-Kravitz et al., 2003), which is likely to prepare the cells for the rapid division

that subsequently occurs. This idea is supported by our TRAP data, which shows greatly

increased telomerase activity in microglia from day 0 to 2 (i.e., nearly 3-fold higher,

compared to controls). Following day 1, GM-CSF treatment induces only maintenance

of the shortest telomeres, as indicated by their continual lengthening until senescence,

while the longest and mean telomeres are allowed to shorten. After day 2, telomerase

activity in GM-CSF-stimulated microglia steadily declines until day 24, then increases

dramatically again to day 32. This supports our Southern blot data, indicating that

telomeres continue to be lengthened until senescence. The elongation of short telomeres

(by 9.1 kb) in GM-CSF-stimulated microglia from day 1 to 16 corresponds with a nearly

3-fold higher increase in telomerase activity during this time. Similarly, short telomeres

increase by 3.1 kb from day 16 to 32, corresponding to an approximate 2-fold increase in

telomerase activity during this time. It is apparent that the rate of elongation of short









telomeres (relative to measured telomerase activity) from day 16 to 32 is lower compared

to day 1 to 16. At day 32, nearly all microglia are likely senescent, and thus may not be

able to lengthen telomeres as efficiently or recruit telomerase as proficiently as on day 2,

suggesting that microglia are less able to utilize available telomerase with increasing time

in culture. The level of telomerase activity is likely enough to maintain the shortest

telomeres while the longest and mean telomeres continually erode. Undergoing rapid

division, as during GM-CSF treatment, appears to enable microglia to rapidly increase

telomere length initially, then on subsequent days to recruit telomerase to the shortest

telomeres, while allowing the longest and mean telomeres to shorten. The opposite is

seen in control microglia, in which only the shortest telomeres erode. Thus, a mechanism

may exist in microglia (following periods of rapid division) that preferentially recruits

limiting amounts of telomerase to maintaining the shortest telomeres while allowing the

longer telomeres to shorten (Ouellette et al., 2000). Telomerase activity may seem

apparent when total protein is isolated and measured following in vitro culturing, yet the

enzyme could be inhibited by a repressor molecule while in vitro or in vivo, which may

or may not be present within the total protein pool during analysis. Additional in vitro or

in vivo molecules may also play critical roles in regulating telomerase activity. Thus, the

pattern of telomerase activity, as determined by in vitro total protein analysis, may not

correlate precisely with or imply telomere maintenance in vitro or in vivo (Ouellette et al.,

1999).

Control microglia are already senescent before day 32 while GM-CSF-stimulated

cells are at or near senescence on the same day, suggesting that GM-CSF treatment may

result in a slightly increased cell life span due to the absence of critically-short telomeres.









Our results suggest that the slightly increased life span of GM-CSF-treated microglia may

be due to their enhanced ability to maintain the shortest telomeres (by telomerase

recruitment), which may result in their delayed entry into senescence, compared to

controls. Perhaps during periods of rapid division, microglia are able to more

proficiently recruit telomerase to the shortest telomeres in an attempt to enable cell

division to occur for a longer period of time prior to entry into senescence. Therefore, the

replicative capacity of rapidly-dividing microglia appears to be greater compared to that

of controls. Microglia were considered senescent by day 32 in control and GM-CSF-

treated groups since they both exhibited decreased proliferation, telomere shortening, and

altered phenotypes (relative to non-senescent dividing microglia). Despite continuous

GM-CSF-stimulation, microglia were unable to maintain a high cell division rate, and

most of the cells had sloughed off from the culture dish by day 32, indicating that they

were no longer viable. Indeed, the yield of DNA was much lower, such that the entire

DNA sample was used for Southern blot analysis for GM-CSF-treated cells on day 32,

whereas only a fraction of the total DNA collected was used for analysis in all other days.

Microglia grown to near-confluence in various-sized culture dishes exhibit increased

telomere attrition with increasing available culture area. Telomere loss likely occurs in

these cultures due to additional cell divisions in microglia that are necessary to reach

near-confluence in the larger-sized culture dishes. There were 4 times as many microglia

plated in the 175 cm2 flask compared to the 9.5 and 21 cm2 dishes. However, the 175

cm2 flask had a culture area that was over 18 times greater than the 9.5 cm2 dish and over

8 times greater than the 21 cm2 dish. Thus, it took approximately 4.5 more divisions per

cell to reach near-confluence in the 175 cm2 flask compared to the 9.5 cm2 flask. These









additional divisions likely account for the decrease in the shortest telomeres evident in

microglia cultured in the 175 cm2 flask compared to the other smaller-sized culture

dishes. Importantly, the shortest telomeres in microglia cultured in the 175 cm2 flask are

up to 3.2 kb shorter, representing an 8.4% decrease, than those found in microglia

cultured in the two smaller-sized dishes. These data demonstrate that the rate of erosion

of the shortest telomeres in microglia is directly proportional to the number of divisions

that a cell undergoes.

Analysis of microglia telomeres by FISH revealed that substantial telomere length

variation occurs both between and within individual chromosomes. This indicates that a

heterogeneous population of telomeres exist within each microglial cell and among sister

chromatids (Bekaert et al., 2002), and further suggests that only a few individual

chromosomes likely reach a critical length first (thereby triggering cell senescence),

assuming uniform telomere shortening with age. The telomeres that are the first to reach

a critical length are likely the first to induce SAGE and TPE on these chromosomes.

Thus, genes (especially those nearest the telomeres) may be up-regulated with age in

microglia due to TPE (Wright and Shay, 1992). Why does so much telomere length

variation exist, especially within the same chromosome? Telomeres on individual

chromosomes may be longer than those on other chromosomes in order to keep gene

regulation and expression constant for that particular area of the chromosome (especially

genes nearest the telomere), or to prevent SAGE and TPE from occurring when telomeres

shorten sufficiently. It would be interesting to determine which genes are located nearest

the telomere on chromosomes which reach a critical length first, since this may provide

clues as to the genetic changes that occur in microglia as they age. Furthermore, this may









suggest an approach (e.g., telomerase over-expression) that would prevent TPE-induced

expression of detrimental genes on chromosomes possessing very short telomeres.

Future studies will permit quantitative FISH (Q-FISH) telomere length measurements, in

which telomere lengths will be quantitated, compared to a standard curve of fluorescence

intensity of plasmids containing known telomeric repeats (Poon et al., 1999). Since sub-

telomeric regions will not be bound by probe (as is the case for Southern blot), this will

allow a more precise measurement of telomere length. Q-FISH will allow a

determination of how much inter- and intra-chromosomal telomere length variation

occurs in microglia and astrocytes with age, will help identify which chromosomes are

the first to reach a critical length and trigger senescence, and if certain chromosomes are

more susceptible to telomere attrition.

Telomeres in astrocytes shorten dramatically during the first few days of in vitro

growth, which is consistent with their rapid division during this time to reach confluence.

A fluctuating pattern of telomere maintenance and attrition follows, which is

accompanied by transient telomerase up-regulation during periods of telomere

maintenance. This fluctuating cycle continues as the cells are maintained through day 32

without passaging, but it is unknown how long this cycle would continue. In astrocytes,

senescence does not occur during these 32 days in culture. This shows that, at least in

vitro, astrocytes have a longer life span than microglia. Perhaps this is because astrocytes

have longer telomeres overall than microglia, and/or because telomerase may be inhibited

in astrocytes while in vivo, but have high activity (due to lack of inhibition) when these

cells are cultured. Small variations exist in telomere lengths between figures 5 and 6,

both of which examine non-passaged astrocytes. This can be attributed to the numerous









experimental parameters that can contribute to the slight variability present from one

experiment to another due to the following: different exposure times (ranging from 1 to 2

hours) during development of the Southern blots, different hybridization times (ranging

from 17 to 19 hours) that the membrane is in contact with the telomere probe, and

especially since different cultures of cells are used for each experiment (each culture has

cells with different rates of mitosis and varying concentrations (about 1.5 rat pup brains

per 175cm2 flask) of whole brain cultures plated initially, which would lead to different

division rates of both astrocytes and microglia prior to their collection, and could

therefore account for this variability).

In passage astrocytes, mean telomere length remains nearly constant while the

longest, and especially the shortest, telomeres fluctuate over time. This suggests that

transient expression of low levels of telomerase may be occurring such that telomerase is

preferentially recruited to the shortest telomeres, while mean telomere length remains

unchanged (Ouellette et al., 2000). Thus, telomerase may be up-regulated (and therefore

maintaining telomere length) during periods of rapid division, as when astrocytes are first

passage. Cultured astrocytes replicate continuously when passage, compared to when

they are plated in 9.5 cm2 dishes and not passage. However, it was noted that with

continual passaging, a slowing of cell division occurred such that it took an increasingly

longer amount of time to reach confluence with successive passages. Once astrocytes

reach confluence in the 9.5 cm2 dish, little replication is likely to occur thereafter. Thus,

non-passaged astrocytes undergo aging in the absence of cell division. However, when

astrocytes are permitted to replicate continually (as when passaged, they maintain

telomeres and even lengthen the shortest ones over time. As in GM-CSF-stimulated






49


microglia, telomeres are initially lengthened when considerable replication is about to

occur. Thus, given the opportunity to continuously replicate, perhaps both microglia

(when exposed to GM-CSF) and astrocytes (when passage) up-regulate telomerase to

compensate for telomere shortening that would otherwise occur. Future studies will

examine telomere length and telomerase activity in microglia and astrocytes in vivo.














CHAPTER 3
TELOMERES SHORTEN WITH AGE IN RAT
CEREBELLUM AND CORTEX IN VIVO

Introduction

Recent data from our laboratory (Flanary and Streit, 2004) has shown that

telomeres shorten with time in cultured microglia, suggesting that these brain cells are

subject to replicative senescence in vivo. Thus, we decided to determine how telomere

length and telomerase activity change in the rat brain with aging. Our results show that

telomere erosion occurs in vivo in the rat cerebellum and cortex with age in the presence

of low levels of steadily increasing telomerase activity.

Materials and Methods

Collection of Rat Cerebellum and Cortex Tissues

Sprague-Dawley rats were maintained at 220C in a controlled 12 hour light/dark

cycle and provided food and water ad libitum. Animals were euthanized by

exsanguination using transcardiac perfusion with phosphate-buffered saline under deep

anesthesia with sodium pentobarbital (50 mg/kg body weight). This method of

euthanasia is consistent with the recommendations of the Panel on Euthanasia of the

American Veterinary Medical Association. Following perfusion, the cerebellum and

cortex were dissected out and frozen prior to DNA and protein isolation.

Determination of Telomere Length

Telomere length was measured as described in Chapter 1.









Determination of Telomerase Activity

Telomerase activity was measured as described in Chapter 1.

Results

Telomeres Shorten With Age in Rat Brain in vivo

Both rat cerebellum and cortex tissue exhibit telomere shortening in vivo from day

21 to 152 (Fig. 3-1, 3-2). The cortex always had shorter telomeres (i.e., longest, mean,

and shortest) than the cerebellum, except on day 152 in the longest telomeres (Fig. 3-3).

In cerebellum, the longest telomeres shortened the most rapidly (loss of 5.6 kb, or 26%),

followed by the mean telomeres (loss of 2.3 kb, or 17%), and the shortest telomeres (loss

of 1.6 kb, or 18%) from day 21 to 152. In the cortex, the mean telomeres shortened most

rapidly (loss of 2.9 kb, or 22%), followed by the longest telomeres (loss of 2.2 kb, or

12%), and the shortest telomeres (loss of 1.0 kb, or 13%) from day 21 to 152. In both

cerebellum and cortex tissues, the shortest telomeres were the slowest to shorten with age

in vivo (Fig. 3-3).

Telomerase Activity in Rat Cerebellum and Cortex

Telomerase activity was measured in both cerebellum and cortex on various days.

With increasing age in vivo, telomerase activity steadily increased in both tissues from

day 21 to 182, with the cerebellum exhibiting the highest activity at all time points

examined (Fig. 3-4, 3-5). From day 21 to 182, telomerase activity increased 28% in the

cerebellum and 11% in the cortex. Further analysis of telomerase activity in cerebellum

and cortex from day 21 to 35 revealed that activity increased slightly in all samples with

age in vivo except for one cerebellum sample from day 28 to 35 (Fig. 3-6, 3-7). Overall,

telomerase activity increased with age in vivo from day 21 to 35 in both cerebellum and

cortex, with the cerebellum exhibiting the highest activity at all time points measured.







From day 21 to 35, telomerase activity increased 12% in the cerebellum and 11% in the
cortex (Fig. 3-8).


woo-a ajai


UUUUUU


23.1 kb-










9.4 kb-


CO




t
0
U


Figure 3-1. Southern blot analysis for measurement of telomere restriction fragment
(TRF) length in rat brain tissue. Genomic DNA was isolated on the indicated
days (day 152 is approximately 5 months of age) from cerebellum and cortex
tissues of two different rats (A and B).

























* Mean 14.0 13.0 12.8 13.8 11.2 11.1 10.4


* Shortest


Figure 3-2.


8.2 9.3 8.0 8.0 7.4 6.8 7.0


TRF length distribution in rat cerebellum and cortex samples on days 21 and
152.


Figure 3-3. Average TRF length in rat cerebellum and cortex tissues on days 21 and 152.
Overall, the cerebellum has longer telomeres than the cortex from day 21 to
152 in all instances (except in longest telomeres on day 152). Telomere
shortening occurs with age in vivo in both rat cerebellum and cortex.
Although only two animals were analyzed for each time point (except cortex
day 152), no overlap of error bars exists between cerebellum and cortex in any
instances.





















rMI Lr f U 00
4;4;-4;6.4 ;6.4-'-4 ; .
0 (, 0 W0 W 0 W 0
UUUUUUUU!P--










b b
"' .'^ ^i" *"


S' 4
Figur 3-4. W4 O







MUM q< W







Figure 3-4. TRAP analysis for telomerase activity in rat brain tissue (days 21 to 182).
Total protein was isolated on the indicated days (days 152 and 182 are
approximately 5 and 6 months of age, respectively) from cerebellum and
cortex tissues. Neg = telomerase-negative control (i.e., cortex A day 21
RNAse-treated extract). Pos = telomerase-positive control.









38-- Cortex A
S36 Cereb A




30


>26

0 20 40 60 80 100 120 140 160 180 200
Time (days)


Figure 3-5. Quantitation of telomerase activity (arbitrary units) in rat cerebellum and
cortex tissues (days 21 to 182). Telomerase activity gradually increases with
age in vivo in both cerebellum and cortex in all instances.






UUUUUUUUUUUQ


i ll




I


Figure 3-6. TRAP analysis for telomerase activity in rat brain tissue (days 21 to 35).
Total protein was isolated on the indicated days from cerebellum and cortex
tissues of two different rats (B and C).










39

38

o37

S36

35

34

S33

32


20 22


26 28 30
Time (Postnatal Day)


Figure 3-7. Quantitation of telomerase activity (arbitrary units) in rat cerebellum and
cortex tissues (days 21 to 35). Telomerase activity gradually increases with
age in vivo in both cerebellum and cortex in all instances (except cerebellum
B from day 28 to 35).


38

S37

S36

S35

S34

33


S32

31


-0- Cortex
-- Cereb


20 22 24 26 28
Time (days)


30 32 34 36


Figure 3-8. Overall, the cerebellum exhibits higher telomerase activity than the cortex
from day 21 to 35.


-0- Cortex B
-- Cereb B
-0- Cortex C
Cereb C





Y^









Discussion

This study, which is the first to report on both telomere length and telomerase

activity in a region-specific manner in rat brain, shows that telomere shortening occurs in

both rat cerebellum and cortex with increasing age in vivo. The telomere shortening is

accompanied by low levels of steadily increasing telomerase activity, which is highest in

the cerebellum in all instances.

Our current findings on telomere length in brain tissues are in apparent conflict

with a recently published study (Cherif et al., 2003). These authors report that while

telomeres shorten with age in rat kidney, liver, lung, and pancreas, no telomere

shortening occurs in rat brain tissue with age (from postnatal day 21 to 15 months of

age). However, their data does indicate that both the longest and shortest telomeres do

shorten in rat brain with age, but not significantly. In addition, the analyses in the study

(Cherif et al., 2003) were based on unspecified brain areas, while our current findings are

in defined brain regions and showed corresponding results (i.e., telomeres shorten in rat

brain with age). Moreover, it appears that the animals used in the study (Cherif et al.,

2003) were not perfused prior to tissue collection, and thus it is likely that the brain tissue

used for telomere length analysis represented a mixture of brain and blood tissue. Blood

contamination of brain tissue could have obscured any brain-specific telomere

shortening. On the other hand, there are two independent studies which have reported

that telomeres do undergo shortening in brain tissue with age. One study performed in

mice demonstrated that telomere shortening does occur in spleen and brain tissue, but not

in liver, testes, or kidney with age from zero to nine months (Prowse and Greider, 1995).

Another study also found that telomeres shorten in mice brain tissue with age from one to

24 months (Coviello-McLaughlin and Prowse, 1997). These three studies (Cherif et al.,









2003; Prowse and Greider et al., 1995; Coviello-McLaughlin and Prowse, 1997) are the

only ones known, at the time of this writing, that have examined telomere length in brain

tissue of rodents with age. Clearly, more work will be required to resolve any conflicting

results that may exist now and to reach a satisfactory conclusion about telomere erosion

in the CNS.

The rate of telomere loss of the longest telomeres in the cerebellum from day 21 to

152 is over two times greater than that in the cortex. This suggests that more cell division

occurs in the cerebellum during this time frame in vivo, compared to the cortex. Our

current study has revealed a difference in telomere lengths between the cerebellum and

cortex by showing that the longest, mean, and the shortest telomeres in the cortex are

consistently shorter than those found in the cerebellum in all but one instance. This

observation may, at first, seem unexpected since cerebellar, unlike cortical, histogenesis

is characterized by the existence of a secondary proliferative zone during late stages of

cerebellar development (Steward, 2000), and one might therefore expect to see shorter

telomeres in the cerebellum. However, our data also show higher telomerase activity in

the cerebellum, compared to the cortex, which thereby may account for the presence of

longer telomeres (despite cell division in the secondary proliferative zone) in the

cerebellum compared to the cortex. Few studies have examined telomerase activity in the

brain. One study (Klapper et al., 2001) reported that telomerase activity is high in mouse

cortex during embryonic development, but sharply decreases during postnatal

development up to three months of age. Likewise, another study (Fu et al., 2000)

reported that telomerase activity is high in rodent neurons during embryonic and early

postnatal development, but then subsequently decreases. Telomerase activity is also high









in rat oligodendrocyte precursor cells, but declines during their differentiation into

oligodendrocytes (Caporaso and Chao, 2001). Our current results indicate that

telomerase activity is lowest on day 21 and steadily increases, albeit slightly, up to six

months of age. In comparison to previous research in this area (Klapper et al., 2001), our

results may or may not coincide. The slight increase in telomerase activity reported in

our study from day 21 to 6 months of age may similar to the same low/decreased levels

as those found in previous studies (Klapper et al., 2001) (after early postnatal

development) if sample sets from both studies were compared on the same gel. It is

nearly impossible to make an accurate comparison between different TRAP gels

(especially between different laboratories), since many factors (e.g., annealing

temperature, number of PCR cycles, protein and primer concentration used, etc.)

influence ladder formation, band intensity, and subsequent telomerase activity levels

(which are expressed in arbitrary units, or as a percentage of the highest levels evident).

The longest telomeres (in cerebellum) and the mean telomeres (in cortex) undergo

the highest rate of telomere loss with age in vivo. However, in addition to the longest

telomeres in the cortex, the shortest telomeres in both tissue types exhibit the slowest rate

of attrition with age. Since telomerase activity slightly increases in both tissue types with

age, this suggests that limiting amounts of telomerase may be preferentially recruited to

the shortest telomeres, while allowing the longest telomeres to shorten (Ouellette et al.,

2000). This data also suggests that the amount of telomerase present may not be enough

to sufficiently compensate for the rate of telomere loss that occurs in the shortest

telomeres with age, since telomeres in both tissue types still shorten with age.









There is high mitotic activity of different, primarily glial, cell types in the

developing rodent brain during the neonatal/postnatal period up to about postnatal day

14. On postnatal day 21, most mitotic activity has ended and the CNS is almost fully

matured. At this time, microglia are probably the only adult cell type remaining in the

postnatal CNS that undergo any appreciable cell division, and they retain their mitotic

ability throughout adult life. Since our analyses began on day 21, we are inclined to think

that most of the telomere erosion subsequent to this day that occurs in both cerebellum

and cortex may be largely attributable to cell division of microglia. Of course, we cannot

exclude the possibility that some of the observed telomere attrition may also be also be

contributed by neural stem cells in the subventricular zones. Future studies employing

tissue from the hippocampus could further illuminate this issue, since neurogenesis

occurs in the hippocampus throughout adult life, but declines with aging (Kuhn et al.,

1996). In addition, some of the observed telomere attrition could also be attributable to

mechanisms not related to cell division, such as oxidative stress (von Zglinicki, 2002), in

postmitotic cells. Future studies in our laboratory will examine telomere length and

telomerase activity in vivo in rat cerebellum and cortex over a longer time period (up to

several years) as well as in acutely isolated microglial cells from young and aged adult

rats.














CHAPTER 4
AXOTOMY INCREASES TELOMERE LENGTH, TELOMERASE ACTIVITY AND
PROTEIN IN AXOTOMY-ACTIVATED MICROGLIA

Introduction

Previous research in our laboratory has demonstrated that telomere shortening and

senescence occurs in cultured rat microglia following periods of prolonged and sustained

mitotic activity induced by GM-CSF (Flanary and Streit, 2004). In contrast, microglia in

vivo undergo short proliferative bursts soon after an acute injury has occurred. This,

together with the fact that microglia produce growth factors and cytokines after injury,

suggests that mitosis affords a mechanism to provide greater numbers of microglial cells

and thus greater trophic support during times of stress (Streit et al., 2000; Streit, 2002a).

However, the mitotic potential of microglia also suggests that these cells have limited

cellular life-spans and thus may rely on proliferation and self-renewal to replace

senescent cells. In order to determine whether neuronal injury-induced microglial

proliferation within a well-defined region of the CNS results in microglial telomere

shortening or cellular senescence, we decided to measure parameters indicative of

telomeric maintenance, such as telomere length and telomerase activity.

Materials and Methods

Rat Facial Nerve Axotomy

Adult Sprague-Dawley rats of both genders were housed at 220C in a controlled 12

hour light/dark cycle and provided food and water ad libitum. Animals were anesthetized

using isofluorane, and the right facial nerve was exposed and transected at the









stylomastoid foramen. Failure to move whiskers on the right side of the face following

recovery from anesthesia was used to verify the success of the axotomy. Animals were

euthanized at 1, 4, 5, 7, and 10 days post-axotomy by exsanguination using transcardiac

perfusion with phosphate-buffered saline (PBS) under deep anesthesia with sodium

pentobarbital (50 mg/kg body weight). This method of euthanasia is consistent with the

recommendations of the Panel on Euthanasia of the American Veterinary Medical

Association. Following perfusion, the entire brain was removed, snap-frozen 2-

methylbutane cooled by liquid nitrogen, and stored at -800C. Individual facial nuclei of

both axotomized (i.e., right) and control (i.e., left) sides were micro-dissected using a

cryostat and stored at -800C prior to DNA and protein isolation.

FACS-Isolation of Rat Microglia from Micro-dissected Facial Nuclei

Animals were exsanguinated under deep sodium pentobarbital anesthesia (50

mg/kg body weight) using ice-cold PBS. The brains were removed and placed in PBS on

ice. Micro-dissection of facial nuclei from the chilled brains was performed immediately

after removal. Nuclei (approximately 7 mg wet weight each) from both axotomized and

non-axotomized sides were collected from six animals at 5 days post-axotomy. Six

nuclei from each side were pooled and processed according to established isolation

protocols (Carson et al., 1998). Fluorescence-activated cell sorting (FACS) analysis was

performed using a FACSVantage SE cell sorter and CellQuest software (BD

Biosciences/Becton Dickinson, San Jose, CA). Monoclonal antibodies used to isolate

microglia during FACS analysis were FITC-conjugated anti-rat CD45 (leukocyte

common antigen), and PE-conjugated anti-rat CD1 lb/c (CR3 complement) (BD









Biosciences/Pharmingen, San Diego, CA), with microglia being identified as the

CD1 b/c high and CD45 low cell population (Ford et al., 1995; Sedgwick et al., 1991).

Determination of Telomere Length

Telomere length was measured as described in Chapter 1.

Determination of Telomerase Activity

Telomerase activity was measured as described in Chapter 1.

Telomerase Western Blot Analysis

Telomerase protein quantity was measured using Western blot analysis and

chemiluminescent detection. Total protein was isolated as described above, and 50 |tg

from each sample, and a telomerase-positive control (rat glioblastoma cell line RG-2),

was separated on an 8% SDS-PAGE gel at 25 mA for 1 hour. Protein was transferred

from the gel to an Immobilon PVDF (polyvinylidene fluoride) membrane (Millipore,

Billerica, MA) using semi-dry transfer at 5 volts for 1 hour. The membrane was blocked

in 5% milk solution in TBST (tris-buffered saline with 0.1% Tween-20) for one hour at

room temperature with shaking, and then incubated overnight in a primary antibody

(rabbit polyclonal anti-telomerase antibody: EST-21A) (Alpha Diagnostics, San

Antonio, TX) at a 1:250 dilution (4 [tg/mL) in 5% milk solution in TBST at 40C with

gentle agitation. A second primary antibody (rabbit polyclonal anti-telomerase antibody:

NB 100-141) (Novus Biologicals, Littleton, CO) was also used in parallel experiments

with the Alpha Diagnostics antibody. Both antibodies yielded similar banding patterns,

however, we found that the Novus Biologicals antibody gave very high background, and

thus used the antibody from Alpha Diagnostics. Following washes in TBST for 4 x 15

min. at room temperature with fast shaking (120 rpm), a secondary antibody (horseradish

peroxidase (HRP)-conjugated anti-rabbit) was applied for one hour at room temperature









with gentle agitation. Following washes in TBST for 4 x 15 min. at room temperature

with fast shaking (120 rpm), chemiluminescent detection was accomplished by

incubating the membrane in the HRP-metabolizing substrate ECL (enhanced

chemiluminescence) (Amersham Biosciences, Piscataway, NJ) for one minute at room

temperature. Following exposure of the membrane to X-omat AR Film (Eastman Kodak

Company, Rochester, NY) in an autoradiography cassette for 1 to 10 min., the film was

developed using a Konica SRX-101A automatic film processor (Konica Minolta,

Mahwah, NJ). A digital image of the autoradiograph was generated by scanning it using

a GS-710 calibrated imaging densitometer (BioRad, Hercules, CA). Quantitation of

telomerase protein quantity was performed using the densitometry computer program

Quantity One (version 4.3.1) (BioRad, Hercules, CA). Following development, the

membrane was re-probed with an anti-GAPDH (glyceraldehyde-3-phosphate

dehydrogenase) antibody as a loading control. The band corresponding to telomerase

protein (catalytic portion) appeared at approximately 60 kDa, while GAPDH appeared at

approximately 40 kDa. Normalized telomerase protein levels were determined by

comparing the average densitometric values of identical samples ran on different gels in

order to make an accurate comparison of protein levels between all samples run on

multiple gels.

Histochemistry

Animals were perfused with 4% paraformaldehyde at 3 days post-axotomy.

Dissected tissues were post-fixed for several days, then rinsed and stored in PBS.

Following paraffin-embedding, serial sections (7.0 |tm thick) were cut on a microtome

(model 2040, Reichert-Jung, Buffalo, NY), mounted onto gelatin-coated slides, and









allowed to dry. Deparaffinization and rehydration were performed by soaking the slides

in xylenes (2 x 15 min.), followed by passage through descending ethanols, including

soaking in 70% ethanol overnight. Lectin histochemistry for the localization of

microglial cells, using the Griffonia simplicifolia B4 isolectin coupled to horseradish

peroxidase (i.e., GSI B4-HRP) (catalog # L5391, Sigma-Aldrich, St. Louis, MO), was

performed as described by Streit (1990). Following development with the peroxidase

substrate diaminobenzadine (DAB), slides were coverslipped with Permount. Selected

sections were counterstained with cresyl violet. Slides were photographed using a Zeiss

Axioskop 2 plus microscope with a RT color Spot camera (model 2.2.1, Diagnostic

Instruments, Inc., Sterling Heights, MI). Photographs were edited using Spot software

(version 3.4.5) (Diagnostic Instruments, Inc., Sterling Heights, MI).

Statistical Analysis of Data

Data was analyzed using the software program InStat version 3.06 (GraphPad, San

Diego, CA). In order to determine whether a statistically significant difference existed

between treatment groups, analysis of variance (ANOVA) was performed using the

Tukey-Kramer multiple comparison test for TRAP assays, while an unpaired t-test was

performed while for Western blot experiments. A p value of >0.05 was considered non-

significant.

Results

Increase in Microglia Surrounding Axotomized Facial Nuclei

Histological examination of sections confirmed prior descriptions of microglial

activation in the axotomized facial nucleus (Fig. 4-1). On the axotomized side, increased

numbers of glial nuclei were apparent and increased lectin staining confirmed that this

was due to greater numbers of microglia. The activated microglia often were gathered







66


around axotomized motor neurons encircling them with their processes. Nuclear staining

with cresyl violet revealed the presence of mitotic figures showing active microglial

proliferation.

Increase in Telomere Length in Axotomized Facial Nuclei

Southern blotting was used to perform telomere length analysis in pooled micro-

dissected rat facial nuclei (Fig. 4-2), with quantitation shown in Fig. 4-3. Telomere

length in axotomized facial nuclei (4 days post-axotomy) increased in the longest (by

21%), mean (by 9%), and shortest (by 10%) telomeres, compared to the control facial

nuclei. This indicates that telomeres lengthen in whole facial nuclei tissue following

axotomy.


- %^ *'b*






a





'* -
^-^.fS

: : < I.
4,- -' ^


S
a


-apt

-'4.
S

rfl

till, 'V

a

4'
S
1~
S


141
A%9



C' ** -



btv
k


Xe. 'a


gut


"'.4




4:





4-


N .. r.... ...a ."* *









10,
i'4


6 1
A *.


4r
.. .


P. .4,
A
~- N'
V
4
e *
I.-. a


Sap
Sr


S .* I'


Figure 4-1. Micrographs of axotomized (A) and control (B) facial nucleus on day 3 post-
axotomy stained with GSI-B4 lectin to identify microglia. Note the mitotic
figure (black arrow, higher magnification shown in inset), and increased
microglial cell numbers on the axotomized side, particularly around injured
motor neurons (red arrows). Scale bar = 25 pm at 40X magnification (A and
B), and 10 atm at 100X magnification (inset).


a iv
fc. ...
h-.-


0
'my


in


*


9"*
9
.7


10







A









23.1 kb a p


C
-w


9.4 kb


6.6 kb w


Figure 4-2. Southern blot analysis for measurement of telomere length in facial nuclei.










50 45.56 EO Longest

37.59
__ 45 ------------~a
S40 -- Shortest
35 ____ ____
30

19.51 17.92

20
15
10 6.48


0 -

FN-Axotomized FN-Control


Figure 4-3. Densitometric quantitation of telomere length in facial nuclei. Genomic
DNA was isolated from 10 pooled axotomized (A) and unlesioned control (C)
facial nuclei on day 4 post-axotomy and probed with a digoxigenin-labelled
telomere specific oligonucleotide (TTAGGG3). Densitometric quantitation
revealed an increase in telomere length in axotomized facial nuclei.

Increase in Telomerase Activity in Axotomized Facial Nuclei

To test whether the observed increases in telomere length were attributable to

similar increases in telomerase activity, TRAP analysis was used for determination of

telomerase activity from day 1 to 10 post-axotomy in individual micro-dissected rat facial

nuclei (Fig. 4-4). Normalized telomerase activity (i.e., of 3 separate TRAP gels with

different samples normalized relative to each other) was significantly increased in the

axotomized facial nuclei (compared to control nuclei) on days 1, 4, 7, and 10 by 8.96%,

68.74%, 104.05%, and 48.25%, respectively (Fig. 4-5). In a non-axotomized animal, no

significant difference in telomerase activity existed between the right and left facial






70

nuclei (Fig. 4-6). These data suggest that one or more different cell types up-regulate

telomerase activity following axotomy.

b00
















0 4IC

:: 6

%-40 -m (boi *:












Figure 4-4. Representative TRAP analysis image used for measurement of telomerase
activity in facial nuclei. Neg = telomerase-negative control; Pos =
telomerase-positive control; IC = internal control.










&100

80

60
t3 S 60
o


g 20

0


0 1 2 3 4 5 Da6
Day


7 8 9 10 11


Figure 4-5. Densitometric quantitation of telomerase activity in facial nuclei. Total
protein was isolated from axotomized (A) or control (C) facial nuclei (day 4
post-axotomy, n = 4). Densitometric quantitation revealed an increase in
mean ( SEM) normalized telomerase activity (arbitrary units) in axotomized
facial nuclei at all timepoints, compared to controls (B). The percent increase
in normalized telomerase activity in axotomized facial nuclei (compared to
controls) following facial nerve axotomy increased the most from day 1 to 4,
and peaked on approximately day 7. Day 0 represents a non-axotomized
animal. Telomerase activity was significantly increased on day 4 (p<0.01),
day 7 (p<0.001), and day 10 (p<0.05) compared to day 1, while day 7 was
higher than day 10 (p<0.01). Total number of animals analyzed on each day
was: day 1 (5), day 4 (6), day 7 (6), day 10 (5).


104.05



S68.74\


/ 48.25 v)



1.1 ,.96































Figure 4-6. Densitometric quantitation of telomerase activity in unoperated facial nuclei.
Absence of variation (p>0.05) in mean ( SD) telomerase activity (arbitrary
units) between left and right non-axotomized facial nuclei from nalve,
unoperated animals (C). Five independent samples were analyzed for both the
left and right facial nuclei. Telomerase activity in the right facial nuclei was
1.42% higher than in left facial nuclei, but was not statistically significant.

Increase in Telomerase Protein Quantity in Axotomized Facial Nuclei

In order to determine if the increases evident in telomerase activity were due to an

increased protein expression, Western blotting was used for determination of telomerase

protein quantity in individual micro-dissected rat facial nuclei (Fig. 4-7). Quantitation of

telomerase protein in rat facial nuclei is shown in Fig. 4-8. Normalized telomerase

protein quantity (i.e., of 3 separate Western blot gels with different samples normalized

relative to each other and to GAPDH) was higher in axotomized facial nuclei (compared

to control nuclei) on day 1 (by 27.68%), 4 (by 45.13%), 7 (by 37.01%%), and 10 (by









103.16%) post-axotomy. This suggests that one or more different cell types up-regulate

telomerase protein following axotomy.


,,,,- ,
"U U to.t% -


--w- -a-


... w w w -*r

gee me


STelomerase


GAPDH


Figure 4-7. Western blot image used for measurement of telomerase protein quantity in
facial nuclei. Total protein was isolated from axotomized (A) or control (C)
facial nuclei on either 1, 4, 7, or 10 days post-axotomy (A). Pos = telomerase-
positive control (RG-2 glioma cells).

35 Axot
Con

25 --
20 25 "4


1 0 --. ....'..................





0 1 2 3 4 5 6 7 8 9 10 11
Days Post-Axotomy

Figure 4-8. Densitometric quantitation of telomerase protein in facial nuclei. An
increase in mean (+ SEM) normalized telomerase protein (arbitrary units) in
axotomized facial nuclei at all timepoints was observed, compared to controls
(B). The percent increase in normalized telomerase protein in axotomized
facial nuclei (compared to controls) following facial nerve axotomy increased
the most from day 1 to 4, and slowly declined from day 4 to 10. Telomerase
protein was significantly increased (p<0.03) on day 4 compared to day 1 in the
axotomized side; all other timepoints are non-significant (p>0.05). Axot =
axotomized facial nuclei; Con = control facial nuclei.


r,,l


I









FACS-Isolation of Microglia from Facial Nuclei

To determine if microglia were the cell type responsible for the increases in

telomerase activity/protein (and hence telomere length), microglia were isolated from six

pooled facial nuclei using CD1 lb/c and CD45 antibodies (Fig. 4-9). Since the samples

contained mixed cell populations, cells were initially sorted based on the RI gate, which

excluded cellular debris (near the lower-left corner) and other undesirable cells (e.g.,

doublet and triplet cells) present elsewhere within the plot (other than RI). Microglia

were identified as the CD1 lb/c high and CD45 low cell population (i.e., the R2 gate).

The parameters of the RI and R2 gates were determined in previous experiments (Carson

et al., 1998). A total of 1,334 microglia (0.46% gated) were isolated from the pooled

axotomized facial nuclei (Fig. 4-9H), whereas only 669 total microglia (0.22% gated)

were isolated from the pooled control facial nuclei (Fig. 4-9J), indicating that at least

twice as many microglia were present in the axotomized facial nucleus compared to the

control nucleus.

Increase in Telomerase Activity in FACS-Isolated Microglia From Axotomized
Facial Nuclei

TRAP analysis was used for determination of telomerase activity in FACS-isolated

microglia from rat facial nuclei (Fig. 4-10). Quantitation of average telomerase activity

in FACS-isolated microglia is shown in Fig. 4-11. Telomerase activity increased by

254.71% in FACS-isolated microglia from axotomized facial nuclei compared to control

facial nuclei. This indicates that microglia within the axotomized facial nucleus exhibit a

large increase in telomerase activity when analyzed separately (i.e., apart from whole

tissue samples), suggesting that they are the cell type mainly responsible for the increases

evident in telomerase activity/protein, and hence telomere length, following axotomy.












0




0


0-



CO
0
t/).0"











A'


0 200 400 600
Forward Scatter


*- -- ..- C c- ... -












", J .'m :
-
P nlm


800 1000 1000


400 600
Forward Scatter


C

0 -

LU







0
0 :
C -


1FF 10


C 0







76



L I
















G 200 400 600 800 1000 100 10 1 10 103 104
Forward Scatter FL CD45 FITC


0i- 5















0 200 400 600 800 1000 I 1 3 4
Forward Scatter FL1 CD45 FITC



Figure 4-9. FACS-isolation of microglia from axotomized and control facial nuclei.
Facial nuclei (axotomized or control) were pooled from rats (5 days post-
axotomy) and subjected to FACS analysis. Antibodies to CD1 lb/c and CD45
were used to isolated microglia. A and B are the negative control (no
antibodies used); C and D are the CD Ilb/c antibody only; E and F are the
CD45 antibody only; G and H are the axotomized facial nuclei; I and J are the
control facial nuclei. Cells within RI were selected and sorted via R2 gating
to isolate microglia.








AC










IC







Figure 4-10. TRAP analysis image used for measurement of telomerase activity in
FACS-isolated facial nuclei.


Figure 4-11. Densitometric quantitation of telomerase activity in FACS-isolated facial
nuclei. Total protein was isolated from FACS-isolated microglia from
axotomized (A) and control (C) facial nuclei (day 5 post-axotomy, six
independent animals, two separate experiments). IC = internal control.
Telomerase activity (arbitrary units) within FACS-isolated microglia from
axotomized facial nuclei on day 5 post-axotomy was increased by 254.71%
compared to the control facial nuclei.









Discussion

We have shown that facial nerve axotomy results in a probable increase in telomere

length and telomerase protein, and a definite increase in telomerase activity in the

axotomized facial nucleus. A definite increase in telomerase activity also occurs in

FACS-isolated activated microglia from the axotomized facial nucleus. These results

support the hypothesis that maintenance and extension of telomere length occurs in

activated microglia that accumulate in axotomized rat facial nuclei. Telomere extension

is likely the result of the observed increases in telomerase protein and activity. Our

findings suggest that dividing microglial cells of the CNS compensate for replication-

induced telomere shortening by up-regulating telomerase.

A total of 10 facial nuclei were pooled for use in Southern blot analysis. This

pooling of tissue samples was necessary due to their small size (approximately 7 mg wet

weight each) and in order to load sufficient quantities of DNA in order to generate a

signal during the Southern blot detection process. Axotomy caused a trend towards an

increase in telomere length (in the longest, mean, and shortest) in the axotomized facial

nuclei, compared to the control facial nuclei. A concomitant increase in both telomerase

activity and telomerase protein was also observed in axotomized facial nuclei. Since

microglia are the only cells known to divide in response to axotomy (Graeber et al.,

1988), these findings suggest that microglia are utilizing telomerase to regulate telomere

length in vivo during periods of high proliferation. We believe that this presumed

increase in microglial telomere length in the axotomized facial nuclei compensates for the

shortening of telomeres that would otherwise occur in the absence of such telomerase

activity. Since FACS-isolated microglia from the axotomized facial nuclei exhibit an

increase in telomerase activity, the increase in telomere length evident within whole









facial nuclei may be attributable to the increased telomerase activity present within

microglia. It remains possible that other cells types may be present which up-regulate

expression, and hence activity, of telomerase following axotomy. Telomerase activity

increased the most from day 1 to 4, and peaked on day 7 post-axotomy, representing a

maximal increase in activity of 104.05%. In view of proliferation data from previous

studies (Graeber et al., 1988; Kreutzberg, 1966; Streit and Kreutzberg, 1988; Svensson et

al., 1994), which showed a peak in microglial proliferation on day 3 post-axotomy, our

findings suggest that the increase in telomerase activity evident within the first 4 days

following axotomy is to prepare microglia for their concomitant proliferative burst

(occurring on days 2 to 4 post-axotomy). Telomerase activity continued to rise from day

4 to 7 as well, indicating the presence of enzymatic activity to lengthen telomeres, if

needed. From 7 to 10 days post-axotomy, telomerase activity declined, possibly since

additional proliferation does not occur during this time period (Graeber et al., 1988;

Kreutzberg, 1996; Streit and Kreutzberg, 1988; Svensson et al., 1994), and thus telomere

length would likely remain stable, and hence telomerase activity would not be required.

Analysis of telomerase activity in naive (i.e., non-axotomized) animals showed no

difference in telomerase activity between the two sides, as expected. The processivity of

the telomerase enzyme (Greider, 1991) (i.e., the quantity of telomeric repeats processed

by the enzyme, as determined by banding pattern on gel), in both whole facial nuclei and

in microglia FACS-isolated from facial nuclei, correlated well with measured telomerase

activity. Lanes possessing an intense initial band (immediately above the internal control

band) always contained a large quantity/intensity of small telomerase products, as well as

a small quantity/intensity of larger products, and are indicative of samples with low









enzyme processivity (i.e., control side of injured facial nucleus). The presence of the

intense initial band indicates that, most of the time, telomerase added only a single

hexanucleotide telomeric repeat onto the end of the TS primer before dissociating. Thus,

most of the measured telomerase activity was contributed by the presence of the first few

initial bands. In these lanes, telomerase apparently is only able to generate small

products and lacks the ability to create larger products, which are clearly present in lanes

with high telomerase processivity (i.e., axotomized side of injured facial nucleus).

There was a two-fold increase in the number of microglia isolated by FACS from

the axotomized facial nuclei (i.e., 1,334) compared to the control facial nuclei (i.e., 669),

which was expected since microglia are known to proliferate in response to axotomy

(Graeber et al., 1988; Kreutzberg, 1996). Since TRAP analysis is a PCR-based assay, it

enabled the measurement of telomerase activity from such small numbers of cells.

FACS-isolated microglia from facial nuclei exhibited an increase in telomerase activity

(by 255%) compared to the control facial nuclei, and thus the increase in telomerase

activity evident within whole facial nuclei tissue samples is most likely attributable to

microglia. Importantly, telomerase activity in FACS-isolated microglia (255% higher in

axotomized side on day 5 post-axotomy) is 270% higher compared to that in whole facial

nuclei tissue samples (69% higher in axotomized side on day 4 post-axotomy). Thus,

when telomerase activity is measured in whole tissue samples, cells other than microglia

which may have low telomerase activity, likely account for the overall decrease in

enzymatic activity.

Few studies have examined telomere length or telomerase activity in the brain, and

most published research in this area has focused on CNS tumors and neural precursor









cells. We have previously shown that telomere shortening occurs in cultured rat

microglia in vitro (Flanary and Streit, 2004), and in rat cerebellum and cortex in vivo in

the presence of low levels of telomerase activity (Flanary and Streit, 2003). One study

reported that telomerase activity is high in mouse cortex during embryonic development,

but sharply decreases during postnatal development up to three months of age (Klapper et

al., 2001). Telomerase activity is also high in rat oligodendrocyte precursor cells, but

declines during their differentiation into mature oligodendrocytes (Caporaso and Chao,

2001). Human neural precursor cells express low levels of telomerase at early passages,

with levels declining to undetectable levels at later passages (greater than 20 population

doublings). In contrast, rodent neural precursor cells express high levels of telomerase at

both early and late passages (Ostenfeld et al., 2000). Telomerase has been found to be

expressed in all brain regions shortly after birth, but becomes restricted to neural stem

cells within the subventricular zone and olfactory bulb in the adult mouse brain

(Caporaso et al., 2003). Likewise, another study reported that telomerase activity is high

in rodent neurons during embryonic and early postnatal development, but decreases

subsequently (Fu et al., 2002). The latter study also found that suppression of telomerase

expression in cultured embryonic hippocampal neurons increased their vulnerability to

apoptosis and excitotoxicity, suggesting that telomerase may play roles other than

telomere maintenance. Induction of telomerase in neurons has been found to exhibit

neuroprotective properties in experimental animal models of neurodegenerative disorders

(Mattson, 2000). Another study reported the induction of telomerase mRNA in cortical

neurons following ischemia (Kang et al., 2004). Thus, telomerase appears to play critical

roles during embryonic development and following brain injury, and it may be









neuroprotective in non-dividing neurons by performing functions unrelated to telomere

length maintenance, such as repair of telomeres damaged by free radicals (von Zglinicki,

2002). The results from our current study in rats also suggest a link between telomerase

and neuroprotection, in that increased telomerase activity may prevent microglial

senescence thereby ensuring sustained glial support of injured neurons. An acute

increase in activated microglia in vivo following axotomy not only serves a beneficial

role, but is also a crucial component of the regenerative process, since microglia divide in

response to the injury, surround and ensheath injured motor neurons, and provide them

with trophic support (Streit, 2002b). However, if multiple bouts of proliferation occurs

(e.g., in response to repeated injury), this may result in an accelerated rate of telomere

shortening in microglia, which may hasten their entry into replicative senescence, and

may thereby limit both the quantity and quality of glial support they are able to provide to

neurons. Recently, astrocytes and microglia have been shown to express telomerase

immunoreactivity in vivo following ischemic or kainite-induced brain injury (Baek et al.,

2004; Fu et al., 2002). We also performed immunohistochemistry for the detection of

telomerase reverse transcriptase, but these experiments were unsuccessful in that no

specific immunoreactivity was observed. The experiments were performed with two

different anti-telomerase antibodies (rabbit anti-telomerase antibody, NB 100-141, from

Novus Biologicals, Littleton, CO; or rabbit anti-telomerase antibody, EST-21A, from

Alpha Diagnostics, San Antonio, TX), using both 4% paraformaldehyde and 10%

formalin fixation with and without antigen retrieval in 0.01 M citrate buffer.

We have shown by Western blot analysis that telomerase protein is increased in the

axotomized facial nucleus compared to the control side. On day 10, there is over a 100%









increase in telomerase activity, which corresponds to, and is likely caused by, the higher

rate of decline of telomerase protein in the control compared to the axotomized side from

day 7 to 10 post-axotomy. It is unknown what causes telomerase protein to decrease at a

slightly faster rate in the control side (relative to the axotomized side) from day 7 to 10

post-axotomy. On day 10, the quantity of telomerase protein is markedly higher than day

1 in the axotomized side, but is slightly lower in controls relative to day 1, suggesting that

telomerase protein (and activity) are necessary in the axotomized side up to at least day

10 post-axotomy. The largest increase detected was on day 4 post-axotomy (45.13%

increase) (p<0.03), followed by day 7 (37.01% increase). This data corresponds well

with the microglial proliferative burst after axotomy in that the largest increase in the

production of telomerase protein is coincident with the maximal number of microglial

cells present 4 days after the axotomy. The increase in telomerase protein also shows

good temporal correspondence with a large increase in telomerase activity suggesting that

increased enzymatic activity may be due to the observed increase in telomerase protein.

Specifically, the increase in telomerase protein preceded the increase in telomerase

activity, since the percent increase in protein peaked on day 4 (45.13% higher in

axotomized facial nuclei), whereas activity peaked on day 7 (104.05% higher in

axotomized facial nuclei). This data suggests that telomerase is translationally-regulated,

at the least. Interestingly, the temporal profiles of telomerase protein followed parallel

patterns in the axotomized and control facial nuclei. Both axotomized and unoperated

control facial nuclei showed increases in telomerase protein from day 1 to 4, and a

decrease from day 4 to 10. This suggests that unilateral axotomy can produce









contralateral effects, an interesting phenomenon that has been noted before but remains

exceedingly difficult to define because of its subtlety and inconsistency.

The findings from our previous studies in vitro (Flanary and Streit, 2004) provided

an impetus to further characterize telomere length and telomerase activity in the facial

nucleus following repeated axotomies to induce continuous mitogenic stimulation. We

found that microglia exposed to continuous mitogenic stimulation with GM-CSF in vitro

show a dramatic increase in telomere length during the first few days post-stimulation,

followed by rapid telomere shortening and senescence. In the current study, a similar

situation is observed in that after microglia undergo a short burst in proliferation

following axotomy, a resultant increase is observed in telomere length. Telomere

shortening may become evident when microglia are subjected to multiple rounds of

proliferative bursts induced by repeated axotomies. These experiments will be the focus

of future studies, and may support the hypothesis that repeated brain injury could lead to

microglial senescence.














CHAPTER 5
ALPHA-TOCOPHEROL (VITAMIN E) INDUCES RAPID, NON-SUSTAINED
PROLIFERATION IN CULTURED RAT MICROGLIA

Introduction

Microglial Activation

Activation of microglial cells is a critical component of the brain's response to

injury. It is particularly prominent after acute lesions when it occurs rapidly and is

characterized, among other things, by a dramatic increase in microglial cell numbers

(Graeber et al., 1988; Kreutzberg, 1996). Granulocyte macrophage-colony stimulating

factor (GM-CSF) has long been known to be an effective microglial mitogen in vitro

(Giulian and Ingeman, 1988; Suzumura et al., 1990), and it has also been implicated as a

stimulator of microglial mitosis after acute injuries (Raivich et al., 1991, 1994). From a

functional point of view, it can be reasonably surmised that the rapid proliferation of

microglial cells shortly after a CNS lesion occurs because greater numbers of these cells

are required to initiate the complex processes of wound healing and tissue repair (Streit et

al., 2000).

Microglial activation is also thought to occur with normal aging and in age-related

neurodegenerative diseases, such as Alzheimer's disease (AD) (McGeer et al., 1987;

Rogers et al., 1988; Streit and Sparks, 1997; Akiyama et al., 2000; McGeer and McGeer,

2001). However, unlike the microgliosis observed after acute CNS injuries, there is no

evidence to show that age- or AD-related microglial activation is accompanied by

increased cell division. Quite to the contrary, there is evidence showing that increased









microglial cell death, as well as microglial structural abnormalities (microglial dystrophy)

are prominent features of the aged and AD brain (Lassmann et al., 1995; Yang et al.,

1998; Jellinger and Stadelmann, 2000; Streit et al., 2004a). The latter observations have

raised the possibility that a loss of microglial cells or of microglial cell function could

contribute to the development of age-related neurodegenerative diseases (Streit, 2002a,

b). Notwithstanding these relatively recent findings, the notion that chronic microglial

activation (often referred to as neuroinflammation) is detrimental and a contributing

factor in AD pathogenesis has become widely accepted and has resulted in the use of

anti-inflammatory regimens as potential treatments (Akiyama et al., 2000). The

neuroinflammation concept has fueled the idea that oxidative stress increases in the aging

brain, in part because activated microglia in vitro produce reactive oxygen species

(Colton and Gilbert, 1987), and it is therefore not surprising that antioxidants have been,

and continue to be, explored as potential anti-aging treatments (Jackson et al., 1988; Sano

et al., 1997; O'Donnell and Lynch, 1998; Milgram et al., 2002; Devi and Kiran, 2004).

Function of Vitamin E

Vitamin E is the most effective lipid-soluble antioxidant in biological membranes,

and it acts to stabilize lipid membranes and prevent propagation of free radical damage

(Halliwell and Gutteridge, 1985). Vitamin E (i.e., alpha-tocopherol) has an organic

structure possessing two aromatic rings and a hydrocarbon tail. Cell surface receptors

exist for the binding and uptake of vitamin E (Kaempf-Rotzoll et al., 2003; Meier et al.,

2003), however, very few studies have examined which specific receptors are responsible

for its uptake. Specific receptor sites have been found for vitamin E on bovine

adrenocortical cells (Kitabchi et al., 1980) human erythrocytes (Kitabchi and

Wimalasena, 1982), and rat hepatocytes (Murphy and Mavis, 1981). Dietary









supplementation of vitamin E has shown benefits for immune cell function (Bendich,

1988; Tengerdy, 1989), as well as cognitive performance and neuroprotection (Socci et

al., 1995; Perrig et al., 1997; Behl and Holsboer, 1998; Joseph et al., 1998; Joseph et al.,

1999; Martin et al., 1999; Behl and Moosmann, 2002; Grundman and Delaney, 2002;

Martin et al., 2002). There are completed and ongoing human clinical trials using

vitamin E as a potential treatment for AD (Sano et al., 1997). The effects of vitamin E on

cultured microglial cells have been studied, and most of this work, consistent with the

idea that microglial activation is a harmful process, has been interpreted to show that

vitamin E suppresses microglial activation (Heppner et al., 1998; Li et al., 2001; Egger et

al., 2001, 2003; Gonzalez-Perez et al., 2002; Godbout et al., 2004; Grammas et al., 2004).

Thus, there appears to be a consensus currently that vitamin E may provide some

neuroprotection by deactivating microglial cells. In the current study, we have

investigated long term effects of vitamin E on primary rat microglial cell cultures with

the goal of determining its effects on cellular aging and proliferation kinetics. The results

reveal that vitamin E is a most potent microglial mitogen that stimulates dramatic

microglial proliferation in vitro. Not unexpectedly, we have also found concomitant

shortening of telomere length and loss of telomerase activity in these cultures.

Materials and Methods

Culturing of Microglia

Microglia were isolated as described in Chapter 1.

Treatment of Microglial Cells

Microglia were treated on day 0 with either 10.2 nM (0.15 [tg/mL) recombinant rat

granulocyte-macrophage colony stimulating factor (GM-CSF) (R&D Systems,

Minneapolis, MN), DL-cL-tocopherol acetate (Sigma-Aldrich, St. Louis, MO; catalog #