- Email: [email protected]
Transcription supports age-related increases of GFAP gene expression in the male rat brain
rr' ELSEVIER
Neuroscience Letters 215 (1996) 107- l 10
Transcription supports age-related increases of GFAP gene expression in the male rat brain T o r u Y o s h i d a , S a r a K. G o l d s m i t h , T o d d E. M o r g a n , D a v i d J. S t o n e , C a l e b E. F i n c h * Neurogerontology Division, Andrus Gerontology Center, and Department (~(Biological Sciences, University of Southern Cali/~)rnia, Los Angeles, CA 90089-0191, USA Received 25 June 1996; revised version received 6 August 1996; accepted 6 August 1996
Abstract During aging, rodent and human brains show progressive increases in the levels of glial fibrillary acidic protein (GFAP) mRNA and protein. The role of transcription was investigated by in situ hybridization, using an intron-containing cRNA probe as a measure of primary GFAP transcripts. We found parallel age-related increases in GFAP intron RNA in the hippocampus, internal capsule, and corpus callosum of 3 versus 24 month old male F344 rats. We conclude that increased transcription supports the age-related increase of GFAP mRNA and protein. GFAP is a unique example of a gene that shows increased expression during aging in contrast to the decreased transcription of certain genes reported in non-neural tissues.
Keywords: Glial fibrillary acidic protein (GFAP); Transcription; Intron; In situ hybridization; Aging; Rat brain
Glial fibrillary acidic protein (GFAP), an intermediate filament protein of astrocytes, shows progressive agerelated increases of expression in all brain regions of those mammals examined. We and others have reported >2-fold age-related increases of GFAP mRNA and protein in the rodent hippocampus and striatum etc. by 24-30 months (mean life span) [3,9,10,13-15,17]. In rodents, the changes are detectable by mid-life (12-18 months), when there are few indications of gross pathological changes in the brain or other organs [21]. Normal human brains also show similar increases of GFAP [15], which suggests that the increased expression of GFAP is among a subset of age changes in the mammalian brain which can not be attributed to specific pathologic conditions [6]. While the increases in GFAP expression roughly parallel the increased numbers of reactive astrocytes in aging rats [4,9,13] and humans [5,12], in situ hybridization data also show a net increase of GFAP mRNA per astrocyte I14]. In first attempts to determine if increased transcription contributes to increased GFAP mRNA we used nuclear
* Corresponding author. Tel.: +1 213 7401758; fax: +l 213 740 0853.
runon assays. These previous studies detected parallel changes in GFAP mRNA and intron-containing transcripts in response to both surgical lesions [11] and glucocorticoids [20]. Despite the sensitivity of these assays, an age trend for increased GFAP transcription rates in cerebral cortex of old rats was not statistically significant [10]. The present approach used in situ hybridization with an intron probe to detect astrocyte subpopulations that might show greater effects of aging. Young (3 months) and old (24 months) male rats (Fl(F344 × BN); purchased from the NIA) were sacrificed under anesthesia (pentobarbital 40 mg/kg). Necropsy did not reveal gross pathological lesions, as expected for these long-lived hybrids at ages below the mean life span of 32 months [21]. Brains were snap-frozen (isopentane, -18°C) and sectioned horizontally at 12 # on a cryostat; sections from both age groups were present on each slide. After fixation by paraformaldehyde and dehydration, in situ hybridization was done according to prior protocols [201 (hybridization buffer 0.75 M NaCI, 50% formamide, 10% dextran sulfate, 0.05 phosphate (pH 7.4); 50°C/18h), followed by RNAse treatment and high stringency washing (0.5 M NaC1, 50% formamide, 0.05 M phosphate (pH 7.4); 63°C) that yielded negligible signals with a sense-strand
0304-3940/96/$12.00 © 1996 Elsevier Science Ireland Ltd. All rights reserved PII $ 0 3 0 4 - 3 9 4 0 ( 9 6 ) 1 2 9 6 6 - 9
108
T. Yoshida et al. / Neuroscience Letters 215 (1996) 107-110
intron probe [20]. Anti-sense 35S-labeled cRNA was transcribed from the pBluescript transcription vector containing a 0.9 kb segment of the rat GFAP intron I (107 cpm//zg cRNA) [ 17]. Slides were dipped in NTB-2 liquid emulsion and exposed for 6 weeks. Sections from eight brains of each age were examined with two independent cycles of counting in three brain regions. Grain density was calculated for 20 cells/region per age by computerized videodensitometry [11] and analyzed by MANOVA (Superanova; Abacus Concepts Inc.). A pilot study with in situ hybridization compared the subcellular distribution of GFAP coding and intron sequences, using a cRNA probe from exon 1 [20], at the same specific activity. As expected, the signal from the intron 1-containing probe was localized to the region of the cell nucleus, whereas that of an exon 1-containing probe was widely distributed throughout the cytoplasm (Fig. 1A).
Three subregions showed marked age changes and were analyzed in detail for grain densities with the intron-containing cRNA probe: the outer molecular layer of the dentate gyrus, the internal capsule, and the corpus callosum. Mean grain density (Fig. IB) showed a significant agerelated increase across all regions (>50%; P < 0.0025, MANOVA). Frequency distributions of grain density classes were non-normal (P < 0.001, all regions; Goodness of Fit test). Each region in old rats had subpopulations of 'hot' astrocytes with high grain densities (bin-size > 16) absent in the young (X2, P < 0.001; Fig. 1C). Both grey and white matter regions (outer molecular layer of the dentate gyrus and myelin rich internal capsule and corpus callosum, respectively) showed astrocytes with increased intron-RNA prevalence; this is consistent with a previous study which showed increased GFAP mRNA per astrocyte in the same regions of hippocampi from old male F344 rats [14]. White matter astrocyte activation during
A
B cDNA containing exon-1
Intron-1
250 ~
Young • 01d []
~,
200 >, t/)
150
t'D
"~ 100 03 50
CC
C
corpus callosum
4 0 - ~ .~
30-
:~
oung
internal capsule
IL\ °'°--I
OML
outer molecular layer
] 30-
•
IC
2015-
20-
1010-
~
1
5O-
0
5
10
15
Grain Density
20
_
0
5
10
__
15
Grain Density
O-
20
0
5
10
15
20
Grain Density
Fig. 1. (A) Photomicrographs of in situ hybridization for GFAP. In situ hybridization for GFAP with intron 1-containing cRNA probe, showing sharp localization of grains over the cell nucleus. In contrast, the exon 1-containing cRNA probe shows diffuse distribution of grains over the cytoplasm. (B) Grain density means for the intron probe were higher in the older brains (24 versus 8 months) in each region **P < 0.001; IC, *P < 0.025. (C) Grain density frequency classes for intron probe showed more intensely labeled cells ('hot astrocytes') in old brains; age difference, CC, P < 0.001; IC, P < 0.001; OML, P < 0.05. CC, corpus callosum; IC, internal capsule; OML, outer molecular layer of the dentate gyrus.
T. Yoshida et al. / Neuroscience Letters 215 (1996) 107-110
aging was also found by immunocytochemistry for GFAP in male rats [3] and in female mice [8]. Failure to detect significant age-related increases in GFAP transcription in runon assays from rat hippocampus and cortex [10] may have resulted from variable amounts of contiguous white matter (and thus variable numbers of hot astrocytes) being included in a given sample during blunt-dissection. This could have resulted in sporadic variation in intron-content in the bulk regions used for nuclear isolation from hot astrocytes in adherent white matter. We conclude that the significant increase of GFAP intron-containing transcripts observed supports the increased expression during aging in cytoplasmic GFAP mRNA and protein. However, there may also be contributions to the age-related increase of GFAP expression from post-transcriptional and post-translational mechanisms, as observed in postnatal brain development and in response to the cytokine tumor necrosis factor (TNF) [12]. Little is known about the turnover of GFAP mRNA and protein at any age; this may be difficult to study because astrocyte subpopulations could have different rates of GFAP biosynthesis, which would be obscured in conventional tracer studies by the need to isolate RNA or protein from bulk regions. Future studies on age changes in GFAP expression should consider the roles of age-related changes in transcription factors [18,22,24]. The GFAP promoter contains a rich array of TRE and other transcriptional response elements that may mediate integration of inflammatory and steroidal mediators of GFAP expression [12]. Age changes in the basal levels of transcription factors include decreases of c-fos mRNA in cerebral cortex of male rats [7], of TRE binding in extracts from hippocampus and cerebral cortex [2], and of Spl binding activity from whole brain [1]. Oxidative stress and other inflammatory processes may be among mechanisms in aging which drive changes in transcription in brain and other tissues. The livers of aging rats show 10-fold increases of nuclear factor K/3 NFr/3 [22,24], a transcription factor that can be induced by oxidative stress and which may also regulate GFAP at a site in its promoter that contains overlapping NF-1 and NFr/3 consensus sequences [12]. Oxidative stress is implicated in the major (>90%) age-related decreases in the transcription of the eC2u-globulin and androgen receptor genes in the aging rat liver [22,24]. Evidence for oxidative stress in the aging brain includes the increased numbers of activated microglia in grey and white matter [4,9,16,19]. Because microglial/macrophage activation can produce reactive oxygen species, astrocyte activation during aging could be a secondary response to oxidative stress from the local microglia. In summary, the effects of aging on gene expression and macromolecular biosynthesis can not be summarized by global statements, because of diverse outcomes of aging that differ between cell types and brain regions [6], as found in other organs [24]. In the case of GFAP, we
109
observed exponential increases of GFAP mRNA during pre- and postnatal development that reach a plateau at about the time of puberty in the rat brain [23], with further increases developing during midlife [15]. Together, these observations on GFAP define a new class in the schedules of gene expression during life history that is characterized by progressive step-wise increases during development and aging. This work was supported by a grant to CEF (AG-7909). [1] Ammendola, R., Mesuraca, M., Russo+ T. and Cimino, F., Spl DNA binding efficiency is highly reduced in nuclear extracts from aged rat tissues, J. Biol. Chem., 267 (1992) 17944-17948. [2] Asanuma, M.+ Kondo, Y., Nishibayashi, S., lwata, E., Nakanashi, T. and Ogawa, N., Age-related changes in composition of transcription factor in the rat brain, Neurosci. Lett., 201 (1995) 127-130. [3] Bronson, R.T., Lipman, R.D. and Harrison, D.E,, Age-related gliosis in the white matter of mice, Brain Res., 609 (1993) 124-128. [4] Gordon, M.N, and Morgan, D.G., Increased GFAP expression in the aged rat brain does not result from increased astrocyte density, Soc. Neurosci. Abstr., 17 (1991) 53 (#26.4). [5] Hansen+ L.A., Armstrong, D.M. and Terry, R.D., An immunohistochemical quantification of fibrous astrocytes in the aging human cerebral cortex, Neurobiol. Aging, 8 (1987) I-6. [6] Johnson, S.A. and Finch, C.E., Changes in gene expression during brain aging, a survey, in E.L. Schneider and J.W. Rowe (Eds.), Handbook of the Biology of Aging+ 4th edn.. Academic Press, San Diego, CA, 1996, pp. 300-327. [7] Kitrabi, E., Bozas, E., Philippidis, H. and Stylianopoulou, F., Agerelated changes in IGF-III and C-los gene expression in the rat brain. Int. J. Dev. Neurosci., I I (1993) 1 9. [8] Kohama, S.G., Goss, J.R., Finch, C.E. and McNeill, T.H., Increases of glial fibrillary acidic protein in the aging female mouse brain+ Neurobiol. Aging, 16 (1995) 105-110. [9] Landfield, P.W, Braun, L.D., Pitier+ T.A.+ Lindsey, J.D. and Lynch, G., Hippocampal aging in rats: a morphometric study of multiple variable in semithin sections, Neurobiol. Aging, 2 (1981) 265-275. [10] Laping, N.J., Teter, B., Anderson+ C., O'Callaghan, J.P., Johnson, S.A. and Finch, C.E., Age-related increases in glial fibrillary acidic protein are not associated with proportionate changes in transcription rates or DNA methylation in the cerebral cortex and hippocampus of male rats, J. Neurosci. Res+, 39 (1994) 710-717. [11] Laping, N.J.+ Morgan, T.E., Nichols, N.R.+ Rozovsky, l., YoungChan. C.S., Zarow, C. and Finch, C.E., Transforming growth factor-~31 induces neuronal and astrocyte genes: tubulin o~i+ glial fibrillary acidic protein, and clusterin. Neuroscience, 58 (1994) 563-572. [12] Laping, N.J., Teter, B., Nichols, N.R., Rozovsky, I. and Finch+ C.E., Glial fibrillary acidic protein: regulation of expression by hormones, cytokines, and growth factors, Brain Pathol. (Special Issue), 4 (1994) 259-274. [13] Lindsey, J.D., Landfield, P.W. and Lynch, G., Early onset and topographical distribution of hypertrophied astrocytes in hippocampus of aging rats: a quantitative study, J. Gerontol.+ 34 (1979) 661671. [14] Major, D.E., Kesslak, J.P., Cotman, C.W., Finch, C.E. and Day, J.R., Life-long dietary restriction attenuates age- related increases in glial fibrillary acidic protein (GFAP) mRNA in the rat hippocampus, Brain Res. (1996) in press. [15] Nichols, N.R.+ Day~ J.R., Laping, N.J., Johnson, S.A. and Finch, C.E., GFAP mRNA increases with age in rat and human brain, Neurobiol. Aging, 14 (1993) 421-429. [16] Nichols, N.R., Finch, C.E. and Morgan, T+E., Age-related increase
110
[17]
[18]
[19] [20]
[21]
T. Yoshida et al. / Neuroscience Letters 215 (1996) 107-110
of TGF-~I mRNA in rat hippocampus is localized to activated microglia, Soc. Neurosci., 21 (1995) 2029. O'Callaghan, J.P. and Miller, D.B., The concentration of glial fibrillary acidic protein increases with age in the mouse and rat brain, Neurobiol. Aging, 12 (1991) 171-174. Papaconstantinou, J., Reisner, P.D., Liu, L. and Kuninger, D.T., Mechanisms of altered gene expression during aging. In E.L. Schneider and J.W. Rowe (Eds.), Handbook of the Biology of Aging, 4th edn., Academic Press, San Diego, CA, 1996, pp. 150-183. Perry, V., Matyszak, M. and Fearn, S., Altered antigen expression of microglia in aged rodent CNS, Glia, 7 (1993) 60-67. Rozovsky, I., Laping, N.J., O'Callaghan, J.P. and Finch, C.E., Transcriptional regulation of GFAP expression by corticosterone in vitro is influenced by the duration of time in culture and by neuron-astrocyte interactions, Endocrinology, 136 (1995) 20662073. Sprott, R.L. and Austad, S.N., Animal models for aging research.
In E.L. Schneider and J.W. Rowe (Eds.), Handbook of the Biology of Aging, 4th edn., Academic Press, San Diego, CA, 1996, pp. 323. [22] Suparkar, P.C., Jung, M.H., Chatterjee, B. and Roy, A.K., Nuclear factor kappa-B functions as a negative regulator for the rat androgen receptor gene and NF-kappa beta activity increases during agedependent desensitization of the liver, J. Biol. Chem., 270 (1995) 837-842. [23] Teter, B., Rozovsky, I., Krohn, K.K., Anderson, C.P., Osterburg, H.H. and Finch, C.E., Methylation of the glial fibrillary acidic protein gene shows novel biphasic changes during brain development, Glia, 17 (1996) 195-205. [24] Van Remmen, H., Ward, W.F., Sabia, R.V. and Richardson, A., Effect of age on gene expression and protein degradation. In E. Masoro (Ed.), Handbook of Physiology, Volume on Aging, Oxford University Press, Oxford, (1994) pp. 171-234.
Neuroscience Letters 215 (1996) 107- l 10
Transcription supports age-related increases of GFAP gene expression in the male rat brain T o r u Y o s h i d a , S a r a K. G o l d s m i t h , T o d d E. M o r g a n , D a v i d J. S t o n e , C a l e b E. F i n c h * Neurogerontology Division, Andrus Gerontology Center, and Department (~(Biological Sciences, University of Southern Cali/~)rnia, Los Angeles, CA 90089-0191, USA Received 25 June 1996; revised version received 6 August 1996; accepted 6 August 1996
Abstract During aging, rodent and human brains show progressive increases in the levels of glial fibrillary acidic protein (GFAP) mRNA and protein. The role of transcription was investigated by in situ hybridization, using an intron-containing cRNA probe as a measure of primary GFAP transcripts. We found parallel age-related increases in GFAP intron RNA in the hippocampus, internal capsule, and corpus callosum of 3 versus 24 month old male F344 rats. We conclude that increased transcription supports the age-related increase of GFAP mRNA and protein. GFAP is a unique example of a gene that shows increased expression during aging in contrast to the decreased transcription of certain genes reported in non-neural tissues.
Keywords: Glial fibrillary acidic protein (GFAP); Transcription; Intron; In situ hybridization; Aging; Rat brain
Glial fibrillary acidic protein (GFAP), an intermediate filament protein of astrocytes, shows progressive agerelated increases of expression in all brain regions of those mammals examined. We and others have reported >2-fold age-related increases of GFAP mRNA and protein in the rodent hippocampus and striatum etc. by 24-30 months (mean life span) [3,9,10,13-15,17]. In rodents, the changes are detectable by mid-life (12-18 months), when there are few indications of gross pathological changes in the brain or other organs [21]. Normal human brains also show similar increases of GFAP [15], which suggests that the increased expression of GFAP is among a subset of age changes in the mammalian brain which can not be attributed to specific pathologic conditions [6]. While the increases in GFAP expression roughly parallel the increased numbers of reactive astrocytes in aging rats [4,9,13] and humans [5,12], in situ hybridization data also show a net increase of GFAP mRNA per astrocyte I14]. In first attempts to determine if increased transcription contributes to increased GFAP mRNA we used nuclear
* Corresponding author. Tel.: +1 213 7401758; fax: +l 213 740 0853.
runon assays. These previous studies detected parallel changes in GFAP mRNA and intron-containing transcripts in response to both surgical lesions [11] and glucocorticoids [20]. Despite the sensitivity of these assays, an age trend for increased GFAP transcription rates in cerebral cortex of old rats was not statistically significant [10]. The present approach used in situ hybridization with an intron probe to detect astrocyte subpopulations that might show greater effects of aging. Young (3 months) and old (24 months) male rats (Fl(F344 × BN); purchased from the NIA) were sacrificed under anesthesia (pentobarbital 40 mg/kg). Necropsy did not reveal gross pathological lesions, as expected for these long-lived hybrids at ages below the mean life span of 32 months [21]. Brains were snap-frozen (isopentane, -18°C) and sectioned horizontally at 12 # on a cryostat; sections from both age groups were present on each slide. After fixation by paraformaldehyde and dehydration, in situ hybridization was done according to prior protocols [201 (hybridization buffer 0.75 M NaCI, 50% formamide, 10% dextran sulfate, 0.05 phosphate (pH 7.4); 50°C/18h), followed by RNAse treatment and high stringency washing (0.5 M NaC1, 50% formamide, 0.05 M phosphate (pH 7.4); 63°C) that yielded negligible signals with a sense-strand
0304-3940/96/$12.00 © 1996 Elsevier Science Ireland Ltd. All rights reserved PII $ 0 3 0 4 - 3 9 4 0 ( 9 6 ) 1 2 9 6 6 - 9
108
T. Yoshida et al. / Neuroscience Letters 215 (1996) 107-110
intron probe [20]. Anti-sense 35S-labeled cRNA was transcribed from the pBluescript transcription vector containing a 0.9 kb segment of the rat GFAP intron I (107 cpm//zg cRNA) [ 17]. Slides were dipped in NTB-2 liquid emulsion and exposed for 6 weeks. Sections from eight brains of each age were examined with two independent cycles of counting in three brain regions. Grain density was calculated for 20 cells/region per age by computerized videodensitometry [11] and analyzed by MANOVA (Superanova; Abacus Concepts Inc.). A pilot study with in situ hybridization compared the subcellular distribution of GFAP coding and intron sequences, using a cRNA probe from exon 1 [20], at the same specific activity. As expected, the signal from the intron 1-containing probe was localized to the region of the cell nucleus, whereas that of an exon 1-containing probe was widely distributed throughout the cytoplasm (Fig. 1A).
Three subregions showed marked age changes and were analyzed in detail for grain densities with the intron-containing cRNA probe: the outer molecular layer of the dentate gyrus, the internal capsule, and the corpus callosum. Mean grain density (Fig. IB) showed a significant agerelated increase across all regions (>50%; P < 0.0025, MANOVA). Frequency distributions of grain density classes were non-normal (P < 0.001, all regions; Goodness of Fit test). Each region in old rats had subpopulations of 'hot' astrocytes with high grain densities (bin-size > 16) absent in the young (X2, P < 0.001; Fig. 1C). Both grey and white matter regions (outer molecular layer of the dentate gyrus and myelin rich internal capsule and corpus callosum, respectively) showed astrocytes with increased intron-RNA prevalence; this is consistent with a previous study which showed increased GFAP mRNA per astrocyte in the same regions of hippocampi from old male F344 rats [14]. White matter astrocyte activation during
A
B cDNA containing exon-1
Intron-1
250 ~
Young • 01d []
~,
200 >, t/)
150
t'D
"~ 100 03 50
CC
C
corpus callosum
4 0 - ~ .~
30-
:~
oung
internal capsule
IL\ °'°--I
OML
outer molecular layer
] 30-
•
IC
2015-
20-
1010-
~
1
5O-
0
5
10
15
Grain Density
20
_
0
5
10
__
15
Grain Density
O-
20
0
5
10
15
20
Grain Density
Fig. 1. (A) Photomicrographs of in situ hybridization for GFAP. In situ hybridization for GFAP with intron 1-containing cRNA probe, showing sharp localization of grains over the cell nucleus. In contrast, the exon 1-containing cRNA probe shows diffuse distribution of grains over the cytoplasm. (B) Grain density means for the intron probe were higher in the older brains (24 versus 8 months) in each region **P < 0.001; IC, *P < 0.025. (C) Grain density frequency classes for intron probe showed more intensely labeled cells ('hot astrocytes') in old brains; age difference, CC, P < 0.001; IC, P < 0.001; OML, P < 0.05. CC, corpus callosum; IC, internal capsule; OML, outer molecular layer of the dentate gyrus.
T. Yoshida et al. / Neuroscience Letters 215 (1996) 107-110
aging was also found by immunocytochemistry for GFAP in male rats [3] and in female mice [8]. Failure to detect significant age-related increases in GFAP transcription in runon assays from rat hippocampus and cortex [10] may have resulted from variable amounts of contiguous white matter (and thus variable numbers of hot astrocytes) being included in a given sample during blunt-dissection. This could have resulted in sporadic variation in intron-content in the bulk regions used for nuclear isolation from hot astrocytes in adherent white matter. We conclude that the significant increase of GFAP intron-containing transcripts observed supports the increased expression during aging in cytoplasmic GFAP mRNA and protein. However, there may also be contributions to the age-related increase of GFAP expression from post-transcriptional and post-translational mechanisms, as observed in postnatal brain development and in response to the cytokine tumor necrosis factor (TNF) [12]. Little is known about the turnover of GFAP mRNA and protein at any age; this may be difficult to study because astrocyte subpopulations could have different rates of GFAP biosynthesis, which would be obscured in conventional tracer studies by the need to isolate RNA or protein from bulk regions. Future studies on age changes in GFAP expression should consider the roles of age-related changes in transcription factors [18,22,24]. The GFAP promoter contains a rich array of TRE and other transcriptional response elements that may mediate integration of inflammatory and steroidal mediators of GFAP expression [12]. Age changes in the basal levels of transcription factors include decreases of c-fos mRNA in cerebral cortex of male rats [7], of TRE binding in extracts from hippocampus and cerebral cortex [2], and of Spl binding activity from whole brain [1]. Oxidative stress and other inflammatory processes may be among mechanisms in aging which drive changes in transcription in brain and other tissues. The livers of aging rats show 10-fold increases of nuclear factor K/3 NFr/3 [22,24], a transcription factor that can be induced by oxidative stress and which may also regulate GFAP at a site in its promoter that contains overlapping NF-1 and NFr/3 consensus sequences [12]. Oxidative stress is implicated in the major (>90%) age-related decreases in the transcription of the eC2u-globulin and androgen receptor genes in the aging rat liver [22,24]. Evidence for oxidative stress in the aging brain includes the increased numbers of activated microglia in grey and white matter [4,9,16,19]. Because microglial/macrophage activation can produce reactive oxygen species, astrocyte activation during aging could be a secondary response to oxidative stress from the local microglia. In summary, the effects of aging on gene expression and macromolecular biosynthesis can not be summarized by global statements, because of diverse outcomes of aging that differ between cell types and brain regions [6], as found in other organs [24]. In the case of GFAP, we
109
observed exponential increases of GFAP mRNA during pre- and postnatal development that reach a plateau at about the time of puberty in the rat brain [23], with further increases developing during midlife [15]. Together, these observations on GFAP define a new class in the schedules of gene expression during life history that is characterized by progressive step-wise increases during development and aging. This work was supported by a grant to CEF (AG-7909). [1] Ammendola, R., Mesuraca, M., Russo+ T. and Cimino, F., Spl DNA binding efficiency is highly reduced in nuclear extracts from aged rat tissues, J. Biol. Chem., 267 (1992) 17944-17948. [2] Asanuma, M.+ Kondo, Y., Nishibayashi, S., lwata, E., Nakanashi, T. and Ogawa, N., Age-related changes in composition of transcription factor in the rat brain, Neurosci. Lett., 201 (1995) 127-130. [3] Bronson, R.T., Lipman, R.D. and Harrison, D.E,, Age-related gliosis in the white matter of mice, Brain Res., 609 (1993) 124-128. [4] Gordon, M.N, and Morgan, D.G., Increased GFAP expression in the aged rat brain does not result from increased astrocyte density, Soc. Neurosci. Abstr., 17 (1991) 53 (#26.4). [5] Hansen+ L.A., Armstrong, D.M. and Terry, R.D., An immunohistochemical quantification of fibrous astrocytes in the aging human cerebral cortex, Neurobiol. Aging, 8 (1987) I-6. [6] Johnson, S.A. and Finch, C.E., Changes in gene expression during brain aging, a survey, in E.L. Schneider and J.W. Rowe (Eds.), Handbook of the Biology of Aging+ 4th edn.. Academic Press, San Diego, CA, 1996, pp. 300-327. [7] Kitrabi, E., Bozas, E., Philippidis, H. and Stylianopoulou, F., Agerelated changes in IGF-III and C-los gene expression in the rat brain. Int. J. Dev. Neurosci., I I (1993) 1 9. [8] Kohama, S.G., Goss, J.R., Finch, C.E. and McNeill, T.H., Increases of glial fibrillary acidic protein in the aging female mouse brain+ Neurobiol. Aging, 16 (1995) 105-110. [9] Landfield, P.W, Braun, L.D., Pitier+ T.A.+ Lindsey, J.D. and Lynch, G., Hippocampal aging in rats: a morphometric study of multiple variable in semithin sections, Neurobiol. Aging, 2 (1981) 265-275. [10] Laping, N.J., Teter, B., Anderson+ C., O'Callaghan, J.P., Johnson, S.A. and Finch, C.E., Age-related increases in glial fibrillary acidic protein are not associated with proportionate changes in transcription rates or DNA methylation in the cerebral cortex and hippocampus of male rats, J. Neurosci. Res+, 39 (1994) 710-717. [11] Laping, N.J.+ Morgan, T.E., Nichols, N.R.+ Rozovsky, l., YoungChan. C.S., Zarow, C. and Finch, C.E., Transforming growth factor-~31 induces neuronal and astrocyte genes: tubulin o~i+ glial fibrillary acidic protein, and clusterin. Neuroscience, 58 (1994) 563-572. [12] Laping, N.J., Teter, B., Nichols, N.R., Rozovsky, I. and Finch+ C.E., Glial fibrillary acidic protein: regulation of expression by hormones, cytokines, and growth factors, Brain Pathol. (Special Issue), 4 (1994) 259-274. [13] Lindsey, J.D., Landfield, P.W. and Lynch, G., Early onset and topographical distribution of hypertrophied astrocytes in hippocampus of aging rats: a quantitative study, J. Gerontol.+ 34 (1979) 661671. [14] Major, D.E., Kesslak, J.P., Cotman, C.W., Finch, C.E. and Day, J.R., Life-long dietary restriction attenuates age- related increases in glial fibrillary acidic protein (GFAP) mRNA in the rat hippocampus, Brain Res. (1996) in press. [15] Nichols, N.R.+ Day~ J.R., Laping, N.J., Johnson, S.A. and Finch, C.E., GFAP mRNA increases with age in rat and human brain, Neurobiol. Aging, 14 (1993) 421-429. [16] Nichols, N.R., Finch, C.E. and Morgan, T+E., Age-related increase
110
[17]
[18]
[19] [20]
[21]
T. Yoshida et al. / Neuroscience Letters 215 (1996) 107-110
of TGF-~I mRNA in rat hippocampus is localized to activated microglia, Soc. Neurosci., 21 (1995) 2029. O'Callaghan, J.P. and Miller, D.B., The concentration of glial fibrillary acidic protein increases with age in the mouse and rat brain, Neurobiol. Aging, 12 (1991) 171-174. Papaconstantinou, J., Reisner, P.D., Liu, L. and Kuninger, D.T., Mechanisms of altered gene expression during aging. In E.L. Schneider and J.W. Rowe (Eds.), Handbook of the Biology of Aging, 4th edn., Academic Press, San Diego, CA, 1996, pp. 150-183. Perry, V., Matyszak, M. and Fearn, S., Altered antigen expression of microglia in aged rodent CNS, Glia, 7 (1993) 60-67. Rozovsky, I., Laping, N.J., O'Callaghan, J.P. and Finch, C.E., Transcriptional regulation of GFAP expression by corticosterone in vitro is influenced by the duration of time in culture and by neuron-astrocyte interactions, Endocrinology, 136 (1995) 20662073. Sprott, R.L. and Austad, S.N., Animal models for aging research.
In E.L. Schneider and J.W. Rowe (Eds.), Handbook of the Biology of Aging, 4th edn., Academic Press, San Diego, CA, 1996, pp. 323. [22] Suparkar, P.C., Jung, M.H., Chatterjee, B. and Roy, A.K., Nuclear factor kappa-B functions as a negative regulator for the rat androgen receptor gene and NF-kappa beta activity increases during agedependent desensitization of the liver, J. Biol. Chem., 270 (1995) 837-842. [23] Teter, B., Rozovsky, I., Krohn, K.K., Anderson, C.P., Osterburg, H.H. and Finch, C.E., Methylation of the glial fibrillary acidic protein gene shows novel biphasic changes during brain development, Glia, 17 (1996) 195-205. [24] Van Remmen, H., Ward, W.F., Sabia, R.V. and Richardson, A., Effect of age on gene expression and protein degradation. In E. Masoro (Ed.), Handbook of Physiology, Volume on Aging, Oxford University Press, Oxford, (1994) pp. 171-234.