- Email: [email protected]
Expression and function of astrocytic gap junctions in aging
Brain Research 901 (2001) 55–61 www.elsevier.com / locate / bres
Research report
Expression and function of astrocytic gap junctions in aging Maria Luisa Cotrina, Qun Gao, Jane H.-C. Lin, Maiken Nedergaard* Department of Cell Biology and Anatomy, and Pathology, New York Medical College, Valhalla, NY 10595, USA Accepted 6 February 2001
Abstract Astrocytic gap junctions have been implicated in a variety of signaling pathways essential to normal brain function. However, no information exists on the prevalence of gap junction channels and their function in the aging brain. Here we have compared the expression of the two most abundant astrocytic gap junction proteins in young and senescent brains and quantified the extent of functional gap junction coupling. The expression level of Cx43 peaked in 7-month-old mice. The relative numbers of Cx43 immunoreactive plaques were 596661, 734662, and 7556114 in 3-, 7-, and 21-month-old mice, whereas plaques size averaged 0.960.1 mm 2 (3 months), 1.360.1 mm 2 (7 months), and 0.760.1 mm 2 (21 months). The expression level of Cx30 was also highest in 7-month-old animals (315649 plaques, size 0.860.07 mm 2 vs. 585651 plaques, size 0.960.1 mm 2 in 3- and 7-month-old mice, respectively), but only 262663 plaques (size 0.460.04 mm 2 ) in 21-month-old mice. Western blot analysis revealed that the content of both Cx43 and Cx30 remained relatively constant at 3, 7, and 21 months. The fluorescence recovery of photobleach technique (FRAP) was used to evaluate coupling in freshly prepared hippocampal slices. Gap junction coupling did not change significantly as a function of aging, but a tendency towards reduced coupling was observed as the animals aged. Average fluorescence recovery after 2 min was 6366% in younger animals, 5965% in adult animals, and 5464% in old brain. These observations indicate that although astrocytic gap junction proteins are maintained at high levels through the entire lifespan of mice, aging is associated with changes in the number and size of both Cx30 and Cx43 gap junction plaques. 2001 Elsevier Science B.V. All rights reserved. Keywords: Glia; Connexins; Brain; Immunohistochemistry; FRAP
1. Introduction Although brain function has been traditionally thought of in terms of neuronal activity, glia cells have been shown within the past few years to actively modulate synaptic transmission. As a result, astrocytes have gained serious consideration as active participants in brain function [19]. Gap junctions are essential to astrocytic signaling functions [3] and it is estimated that more than 50 000 gap junction channels interconnect each astrocyte to its neighbors [22]. Astrocytic gap junctions are mainly composed of connexin 43 (Cx43), although Cx30, Cx40 and Cx45 are also expressed in lower amounts [9]. The other cell types in brain, including neurons and oligodendrocytes are poorly coupled in the adult [9], although a recent report has suggested that interneuronal coupling is more widespread than previously recognized [1]. The roles of astrocytic gap junctions continue to increase, but include the distribution *Corresponding author. Tel.: 11-914-5944-111; fax: 11-914-5944453.
of both metabolic substrates and products within the brain [20], the redistribution of potassium ions after neuronal electrical activity [17], and cell–cell communication via calcium waves [19]. In addition, recent studies have shown that connexin proteins serve other roles that do not necessarily require the formation of gap junction channels. Cx-expression regulates ATP release from glia [7], modulates cytoskeletal organization of transfected cells [6], influences neuronal differentiation [8], modulates cell–cell adhesion [4], and increases cellular migration [11]. Moreover, it has been recently demonstrated that ischemic astrocytes remain coupled during cell death both in vivo and in vitro [5]. Indeed, gap junction-mediated signaling can propagate and amplify brain injury to include otherwise viable glial cells [15]. Collectively, these studies indicate a profound involvement of connexin proteins in both normal and aberrant brain function. To date, aging-related modulations of gap junction expression and coupling have not been evaluated. In this study, we have investigated the level of expression and function of the two most abundant gap junction proteins of
0006-8993 / 01 / $ – see front matter 2001 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 01 )02258-2
56
M.L. Cotrina et al. / Brain Research 901 (2001) 55 – 61
astrocytes, Cx43 and Cx30, in an attempt to define how aging modulates astrocytic coupling.
CDCF diacetate, is restricted to astrocytes in freshly prepared slices [12,13].
2.4. Immunohistochemistry 2. Methods
2.1. Animals and housing conditions Three groups of C57B1 / 6 mice were studied: 3-monthold (young group), 7-month-old (adult group) and 21month-old mice (aged group). The mice were obtained from the aging colony of the National Institute of Aging and housed in standard cages with accessible food pellets and water and kept in a 12 h light–dark cycle.
2.2. Preparation of brain slices The animals were killed by decapitation while in deep pentobarbital anesthesia (80 mg / kg), the brains were removed and 300 mm coronal sections were cut on a vibratome (TPI, St. Louis, MO, USA) under low-calcium, high-magnesium artificial cerebrospinal fluid (ACSF) containing (in mM): 126 NaCl, 2.5 KCl, 1.25 NaH 2 PO 4 , 10 MgSO 4 , 0.5 CaCl 2 and 10 glucose [12]. Slices were allowed to recover in standard ACSF (126 mM NaCl, 2.5 mM KCl, 1.25 mM NaH 2 PO 4 , 2 mM MgCl 2 , 2 mM CaCl 2 , 10 mM glucose and 26 mM NaHCO 3 ) oxygenated with 5% CO 2 and 95% O 2 for 1–2 h.
2.3. Fluorescence recovery after photobleaching ( FRAP) assay Slices prepared as described above were loaded with CDCF (dicarboxydichlorofluorescein diacetate), a gap junction- and cell-membrane-permeable dye that becomes membrane-impermeable after de-esterification. Brain slices were incubated in 1 M of the dye dissolved in ACSF. After several washes in the same solution in the absence of CDCF (to allow for complete de-esterification), the slices were placed in the stage of an inverted microscope attached to a Bio-Rad 1000 confocal microscope (Hercules, CA, USA). After excitation with the 488 nm line of the krypton / argon laser, a baseline fluorescence was collected using a 403 objective. Then, the area of laser scanning was reduced and the fluorescence signal bleached using 103 zoom and full power of laser. After three to five scans, the normal recording configuration was reestablished and refill was recorded for 40 s [11,15]. All studies were performed using optical sections 30 mm below the surface to ensure that the cells of interest had not been damaged during preparation. The FRAP technique does not discriminate between cell types. However, when utilized in adult brain tissue, fluorescence refill primarily reflects astrocytic coupling, as (1) only astrocytes are coupled to a significant extent in adult brain and (2) de-esterification of fluorescence indicators, including
A rabbit anti-mouse Cx43 (kindly provided by A. Lau), and a rabbit anti-rat Cx30 antibodies (Zymed, catalog No. 71-2200) were used. Eight mm cryostat sections of fresh brain tissue were fixed with methanol [15,14]. After overnight incubation with a 1:1000 dilution of anti-Cx43 antibody and a 1:250 dilution for anti-Cx30 antibody, detection was performed by applying a goat anti-rabbit secondary antibody conjugated to FITC (Jackson). Slices were mounted in Slow Fade (Molecular Probes) and examined by confocal microscopy (MRC1000; Bio-Rad) through a Nikon Diaphot inverted scope. Image analysis was performed using the Q500MC program from Leica. Analayiss of variance (ANOVA) was used to compare mean values whereas Student’s t-test was applied to determine which groups significantly differed. Number of Cx-immunoreactive plaques per 0.2 mm 2 was quantified in a minimum of 5 fields in each animal.
2.5. Western blot analysis Brain slices were homogenized by a tissue grinder in 10 mM Tris buffer and lysed in 2% sodium dodecyl sulphate (SDS) buffer [10]. After determination of the protein content by the Bradford assay (Bio-Rad) with bovine serum albumin (BSA) as standard, samples were diluted with 33 sodium dodecyl sulfate (SDS) gel-loading buffer. For detection of connexins, 50 mg of homogenate protein was resolved on 10% SDS–polyacrylamide gel electrophoresis (PAGE) gels, transferred to nitrocellulose membranes, blocked overnight at 48C in TBST (Tris, NaCl and Tween-20) with 3% BSA and incubated for 2 h at room temperature with the polyclonal anti-Cx43 1:1000) or the anti-Cx30 1:250) antibody. After repeated washes, an horesradish peroxidase (HRP)-conjugated goat anti-rabbit IgG was applied and antibody binding visualized by using an enhanced chemiluminescence kit (Amersham).
3. Results
3.1. Connexin-43 expression is higher in slices from senescent than from young hippocampus Immunohistochemical analysis of hippocampal sections from young adult mice (3 months), adult (7 months) and from old mouse brain (21 months) revealed that Cx43, the major gap junction protein of astrocytes, was organized in immunoreactive plaques as expected (Fig. 1). In accordance with earlier reports [16,15], the Cx-plaques were localized primarily to the plasma membrane, a pattern identical to the one observed in dissociated cultures of
M.L. Cotrina et al. / Brain Research 901 (2001) 55 – 61
57
Fig. 1. The expression of gap junction proteins increases as the brain matures. Immunohistochemistry of hippocampal sections from 3-, 7- and 21-month-old mice with antibodies directed against (A) Cx43 or (B) Cx30. Both the expression of Cx43 and Cx30 peaked in 7-month-old mice. Scale bar, 5 mm. (C) Bar histogram of the overall immunoreactivity of Cx43 and Cx30 in hippocampus of mice of the same ages as illustrated in A and B. Rel. immunoreactivity indicates mean number of plaques per 0.2 mm 2 . * Cx43, significantly different from 3-month-old mice. ** Cx30, significantly different from 3-month-old mice. P,0.05 by t-test assay.
cortical astrocytes [5]. Interestingly, higher levels of Cx43 immunoreactivity were evident in adult vs. young animals. Although the number of immunoreactive Cx43 plaques detected in the older animals was not significantly higher than in the younger animals (596661 plaques / 0.2 mm 2 in 3-month-old mice vs. 734662 / 0.2 mm 2 and 7556114 / 0.2 mm 2 plaques in 7- and 21-month-old mice, respectively; P.0.1 Student’s t-test), the size of the gap junction plaques increased as a function of age (0.960.1 and 1.360.1 mm 2 in 3- and 7-month-old mice, respectively;
P,0.02 Student’s t-test). However, the increase in the number of plaques was inversely proportional to the size of the plaques in the senescent brain (0.760.1 mm 2 , Fig. 1). A similar picture emerged when we analyzed the expression pattern of Cx30, which reaches maximal expression in the adult rat brain [14]. High levels of Cx30 were also detected in the brain of adult mice. In fact, a twofold increase in connexin expression was evident in adult compared to younger animals (315649 plaques / 0.2 mm 2 , size 0.860.07 mm 2 vs. 585651 / 0.2 mm 2 plaques,
58
M.L. Cotrina et al. / Brain Research 901 (2001) 55 – 61
size 0.960.1 mm 2 in 3- and 7-month-old mice, respectively; P,0.001 Student’s t-test) but only 262663 / 0.2 mm 2 plaques (size 0.460.04 mm 2 ) were detected in the brain of old mice. These data indicate that there is an overall loss of immunoreactivity against both Cx43 and Cx30 in older rats. Specifically, the size of the Cx43 immunoreactive plaques was reduced, whereas there was a decrease in both the number and size of the Cx30 plaques in old mice.
3.2. Gap junctions are functional in the aging brain Measurement of functional coupling in intact mouse brain slices using FRAP demonstrated that functional coupling was not significantly altered as a function of aging (Fig. 2). Average recovery of fluorescence after 2 min was 6366% in younger animals, (three mice, 3 months, a total of 13 FRAP experiments), 5965% in adult animals (two mice, 7 months, a total of 16 FRAP experiments) and 5464% in old brain (two mice, 19 and 21 months, a total of 15 FRAP experiments). Although gap junction permeability did not significantly change, an overall tendency to decreased coupling was observed as the animal aged.
3.3. Western blot analysis of Cx43 and Cx30 The results presented above document that the number and / or size of Cx43 and Cx30 immunoreactive plaques decline during the aging process in hippocampus. We next tested whether the decline in Cx-plaques correlated with the amount of connexin protein present in brain. To this end, we used Western blot analysis of whole brain extracts. Fig. 3 shows that the levels of both Cx43 and Cx30 remained relatively constant throughout the lifespan of the mice studied. Thus, the decline in Cx43 and Cx30 immunoreactive plaques does not result from a general decrease in the pool of Cx-proteins, but rather from redistribution. Of note, in accordance with earlier reports, we only observed the unphosphorylated 41 kDa band and not the 43 kDa phosphorylated band [10]. Brain tissue contains endogenous phosphatases that rapidly unphosphorylate Cx43 during tissue preparation, unless brain metabolism is rapidly inactivated by cranial high-energy microwave irradiation or inclusion of phosphatase inhibitors [10].
4. Discussion Little information exists with regard to the aging process of astrocytes and no studies have explored so far how well these cells communicate in the senescent brain. Here we have mapped the expression levels of the most abundant astrocytic gap junction proteins, Cx43 and Cx30, and analyzed the extent of functional coupling in young versus
aged brain tissue. Our data show that the levels of Cx43, a gap junction protein that is expressed from very early in development in most astrocytes [16], decreased in aging. Aging was associated with a reduction in the average size of Cx43 immunoreactive plaques, whereas the number of plaques remained essentially unchanged in senescent brains (21-month-old mice). A slightly different pattern of age-induced modulation of another member of the Cxfamily, Cx30, was observed. Cx30, a gap junction protein with a late onset of expression in the adult murine brain [14], is indeed expressed in young adult mice (3-monthold). Its expression is increased in adult mice, but decreases again as the brain ages. In fact, both the number of Cx30-immunoreactive plaques, as well as the size of the plaques were clearly reduced in old tissue. In rat hippocampus, aging has been associated with an increase in the number and size of astrocytes [2]. It is therefore not possible to estimate the relative number of Cx-plaques per astrocytes as a function of aging. We used the FRAP technique to evaluate functional gap junction coupling in freshly prepared hippocampal slices. The logic for using this approach was that intracellular injection of gap-junction-permeable tracers becomes increasingly difficult as a function of the age of the mice brain (our unpublished observations). We were concerned that the procedures associated with tracer injection would affect the outcome of the analysis and preferentially would decrease tracer spread in the aging brain. We therefore chose to analyze gap junction coupling using the noninvasive FRAP technique. The fact that coupling remained relatively high in the older animals suggests that our measurements accurately reflect that astrocytes remain coupled during the aging process. It is interesting to note that the size of the gap junction plaques was reduced in aged brain. The reduction in plaque size was not associated with a significant decrease in functional coupling, suggesting that the reduction in plaque size did not significantly impair channel function. However, as the brain ages, a reduction in plaque size may be indicative of a modulation, rather than a loss of connexin function. A similar dissociation between reduction of connexin levels but constant intercellular communication has been recently observed in mammary epithelial cells undergoing changes in their activation state [18]. In this regard, alternative roles for connexins that do not require the formation of gap junction channels have been described in rat astrocytic cultures [6,8]. In ischemic brain tissue, aggregation, rather than disassembly of gap junction plaques was observed in astrocytes during the process of cell death [5]. In this study, the dramatic increase in plaque size was associated with a clear reduction in the degree of phosphorylated Cx43 and a consequent loss of channel permeability. Alternatively, the modulation we observe in the number and size of the gap junction plaques might represent a reorganization of the plaques within the network of autocellular contacts of the astrocytes. Indeed,
M.L. Cotrina et al. / Brain Research 901 (2001) 55 – 61
59
Fig. 2. Gap junction functionality is preserved in brains from old mice. The fluorescence recovery after photobleach technique (FRAP) was used to quantify gap junction function. Top panels illustrate hippocampal slices from 3-month-old (left) and 21-month-old mice (right) loaded with the gap junction permeable tracer CDCF. Middle panel illustrates the area selected for laser bleaching (red rectangle). Lower panels show refill of fluorescence 2 min after laser bleach (arrow). Bar histogram maps the percentage of recovery after photobleach in 3-, 7- and 21-month-old mice. Fluorescence recovery does not differ significantly between the different ages, although a trend towards a decrease is evident in older mice.
60
M.L. Cotrina et al. / Brain Research 901 (2001) 55 – 61
References
Fig. 3. Western blot analysis of gap junction expression in aging. Lane 1, 3-month-old mice; lane 2, 7-month-old mice; lane 3, 21-month-old mice. The expression levels of both Cx43 and Cx30 remained relatively constant.
reflexive gap junctions have been reported to represent a considerable amount of the total astrocytic gap junctions [21]. Although the meaning of reflexive gap junctions is still unknown, changes in the organization of these connections would not necessarily affect gap junction permeability. In conclusion, although astrocytic gap junction plaques are reduced in the brains of old mice compared to young adults, functional coupling remains high. The reorganization of Cx-immunoreactive plaques in the senescent brain, which is expressed as a redistribution in plaque size (Cx43) or a reduction of immunoreactive plaques (Cx30), suggests that functions other than channel permeability are modulated in aging. The observation provides an experimental framework for the manipulation of gap junction-dependent glial calcium signaling and, by extension, other connexin-dependent neural processes that may be impaired in aging.
Acknowledgements This study was supported by NINDS / NIH grants NS30007 and NS35011. M.N. is an Established Investigator sponsored by The American Heart Association.
[1] V. Alvarez-Maubecin, F. Garcia-Hernandez, J.T. Williams, E.J. Van Bockstaele, Functional coupling between neurons and glia, J. Neurosci. 20 (2000) 4091–4098. [2] F. Amenta, E. Bronzetti, M. Sabbatini, J.A. Vega, Astrocyte changes in aging cerebral cortex and hippocampus: a quantitative immunohistochemical study, Microsc. Res. Tech. 43 (1998) 29–33. [3] A. Araque, V. Parpura, R.P. Sanzgini, P.G. Haydon, Tripartite synapses: glia, the unacknowledged partner, Trends Neurosci. 22 (1999) 208–215. [4] G. Arcuino, S.A. Goldman, M.L. Cotrina, S. Liu, V. Fried, J.H.-L. Lin, M. Nedergaard, Connexin can direct selective cell–cell interactions, Soc. Neurosi. Abstr. 25 (1999) 600.11. [5] M.L. Cotrina, J. Kang, J.H.-C. Lin, E. Bueno, T.W. Hansen, L. He, Y. Liu, M. Nedergaard, Astrocytic gap junctions remain open during ischemic conditions, J. Neurosci. 18 (1998) 2520–2537. [6] M.L. Cotrina, J.H.-C. Lin, M. Nedergaard, Cytoskeletal assembly and ATP release regulate astrocytic calcium signaling, J. Neurosci. 18 (1998) 8794–8804. [7] M.L. Cotrina, J.H.-C. Lin, A. Alves-Rodrigues, S. Liu, J. Li, H. Azmi-Ghadimi, J. Kang, C.C.G. Naus, M. Nedergaard, Connexins regulate calcium signaling by controlling ATP release, Proc. Natl. Acad. Sci. USA 95 (1998) 15735–15740. ´ [8] M.L. Cotrina, J.H.-C. Lin, J.C. Lopez-Garcia, C.C.G. Naus, M. Nedergaard, Connexin-mediated glia signaling, J. Neurosci. 20 (2000) 2835–2844. [9] R. Dermietzel, D. Spray, From neuro-glue (‘nervenkitt’) to glia: a prologue, Glia 24 (1998) 1–7. [10] M.Z. Hossain, L.J. Murphy, E.L. Hertzberg, J.I. Nagy, Phosphorylated forms of connexin 43 predominate in rat brain: demonstration by rapid inactivation of brain metabolism, J. Neurochem. 62 (1994) 2394–2403. [11] G.Y. Huang, E.S. Cooper, K. Waldo, M.L. Kirby, N.B. Gilula, C.W. Lo, Gap junction mediated cell–cell communication modulates mouse neural crest migration, J. Cell Biol. 143 (1998) 1725–1734. [12] J. Kang, L. Jiang, S.A. Goldman, M. Nedergaard, Astrocyte-mediated potentiation of inhibitory synaptic transmission, Nat. Neurosci. 1 (1998) 683–692. [13] J. Kang, M. Nedergaard, Imaging astrocytes in acute brain slices, in: R. Yuste, F. Lanni, A. Konnerth (Eds.), Imaging Structure and Function in the Nervous System, Cold Spring Harbor Laboratory Press, 2000, pp. 42.1–42.11. [14] P. Kunzelmann, W. Schroder, O. Traub, C. Steinhauser, R. Dermietzel, K. Willecke, Late onset and increasing expression of the gap junction protein connexin30 in adult murine brain and long-term cultured astrocytes, Glia 25 (1999) 111–119. [15] J.H.-C. Lin, H. Weigel, M.L. Cotrina, S. Liu, E. Bueno, A.J. Hansen, T.W. Hansen, S. Goldman, M. Nedergaard, Gap-junction-mediated propagation and amplification of cell injury, Nat. Neurosci. 6 (1998) 494–500. [16] B. Nadarajah, D. Thomaidou, W.H. Evans, J.G. Parnavelas, Gap junctions in the adult cerebral cortex: regional differences in their distribution and cellular expression of connexins, J. Comp. Neurol. 376 (1996) 326–342. [17] B.R. Ransom, Do glial gap junctions play a role in extracellular ion homeostasis, in: D.C. Spray, R. Dermietzel (Eds.), Gap Junctions in the Nervous System, R.R. Landes, Austin, TX, 1996, pp. 160–173. [18] M.A. Sia, T.L. Woodward, J.D. Turner, D.W. Laird, Quiescent mammary epithelial cells have reduced Cx43 but maintain a high level of gap junction intercellular communication, Dev. Genet. 24 (1999) 111–122. [19] S.J. Smith, Neuromodulatory astrocytes, Curr. Biol. 4 (1994) 807– 810. [20] A. Tabernero, C. Giaume, J.M. Medina, Endothelin-1 regulates glucose utilization in cultured astrocytes by controlling intercellular communication through gap junctions, Glia 16 (1996) 187–195.
M.L. Cotrina et al. / Brain Research 901 (2001) 55 – 61 [21] J.R. Wolff, K. Stuke, M. Missler, H. Tytko, P. Schwarz, A. Rohlmann, T.I. Chao, Autocellular coupling by gap junctions in cultured astrocytes: a new view on cellular autoregulation during process formation, Glia 24 (1998) 121–140.
61
[22] T. Yamamoto, E. Ochalski, E. Hertzberg, J. Nagy, On the organization of astrocytic gap junctions in rat brain as suggested by LM and EM immunohistochemistry of connexin43 expression, J. Comp. Neurol. 302 (1990) 853–883.
Research report
Expression and function of astrocytic gap junctions in aging Maria Luisa Cotrina, Qun Gao, Jane H.-C. Lin, Maiken Nedergaard* Department of Cell Biology and Anatomy, and Pathology, New York Medical College, Valhalla, NY 10595, USA Accepted 6 February 2001
Abstract Astrocytic gap junctions have been implicated in a variety of signaling pathways essential to normal brain function. However, no information exists on the prevalence of gap junction channels and their function in the aging brain. Here we have compared the expression of the two most abundant astrocytic gap junction proteins in young and senescent brains and quantified the extent of functional gap junction coupling. The expression level of Cx43 peaked in 7-month-old mice. The relative numbers of Cx43 immunoreactive plaques were 596661, 734662, and 7556114 in 3-, 7-, and 21-month-old mice, whereas plaques size averaged 0.960.1 mm 2 (3 months), 1.360.1 mm 2 (7 months), and 0.760.1 mm 2 (21 months). The expression level of Cx30 was also highest in 7-month-old animals (315649 plaques, size 0.860.07 mm 2 vs. 585651 plaques, size 0.960.1 mm 2 in 3- and 7-month-old mice, respectively), but only 262663 plaques (size 0.460.04 mm 2 ) in 21-month-old mice. Western blot analysis revealed that the content of both Cx43 and Cx30 remained relatively constant at 3, 7, and 21 months. The fluorescence recovery of photobleach technique (FRAP) was used to evaluate coupling in freshly prepared hippocampal slices. Gap junction coupling did not change significantly as a function of aging, but a tendency towards reduced coupling was observed as the animals aged. Average fluorescence recovery after 2 min was 6366% in younger animals, 5965% in adult animals, and 5464% in old brain. These observations indicate that although astrocytic gap junction proteins are maintained at high levels through the entire lifespan of mice, aging is associated with changes in the number and size of both Cx30 and Cx43 gap junction plaques. 2001 Elsevier Science B.V. All rights reserved. Keywords: Glia; Connexins; Brain; Immunohistochemistry; FRAP
1. Introduction Although brain function has been traditionally thought of in terms of neuronal activity, glia cells have been shown within the past few years to actively modulate synaptic transmission. As a result, astrocytes have gained serious consideration as active participants in brain function [19]. Gap junctions are essential to astrocytic signaling functions [3] and it is estimated that more than 50 000 gap junction channels interconnect each astrocyte to its neighbors [22]. Astrocytic gap junctions are mainly composed of connexin 43 (Cx43), although Cx30, Cx40 and Cx45 are also expressed in lower amounts [9]. The other cell types in brain, including neurons and oligodendrocytes are poorly coupled in the adult [9], although a recent report has suggested that interneuronal coupling is more widespread than previously recognized [1]. The roles of astrocytic gap junctions continue to increase, but include the distribution *Corresponding author. Tel.: 11-914-5944-111; fax: 11-914-5944453.
of both metabolic substrates and products within the brain [20], the redistribution of potassium ions after neuronal electrical activity [17], and cell–cell communication via calcium waves [19]. In addition, recent studies have shown that connexin proteins serve other roles that do not necessarily require the formation of gap junction channels. Cx-expression regulates ATP release from glia [7], modulates cytoskeletal organization of transfected cells [6], influences neuronal differentiation [8], modulates cell–cell adhesion [4], and increases cellular migration [11]. Moreover, it has been recently demonstrated that ischemic astrocytes remain coupled during cell death both in vivo and in vitro [5]. Indeed, gap junction-mediated signaling can propagate and amplify brain injury to include otherwise viable glial cells [15]. Collectively, these studies indicate a profound involvement of connexin proteins in both normal and aberrant brain function. To date, aging-related modulations of gap junction expression and coupling have not been evaluated. In this study, we have investigated the level of expression and function of the two most abundant gap junction proteins of
0006-8993 / 01 / $ – see front matter 2001 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 01 )02258-2
56
M.L. Cotrina et al. / Brain Research 901 (2001) 55 – 61
astrocytes, Cx43 and Cx30, in an attempt to define how aging modulates astrocytic coupling.
CDCF diacetate, is restricted to astrocytes in freshly prepared slices [12,13].
2.4. Immunohistochemistry 2. Methods
2.1. Animals and housing conditions Three groups of C57B1 / 6 mice were studied: 3-monthold (young group), 7-month-old (adult group) and 21month-old mice (aged group). The mice were obtained from the aging colony of the National Institute of Aging and housed in standard cages with accessible food pellets and water and kept in a 12 h light–dark cycle.
2.2. Preparation of brain slices The animals were killed by decapitation while in deep pentobarbital anesthesia (80 mg / kg), the brains were removed and 300 mm coronal sections were cut on a vibratome (TPI, St. Louis, MO, USA) under low-calcium, high-magnesium artificial cerebrospinal fluid (ACSF) containing (in mM): 126 NaCl, 2.5 KCl, 1.25 NaH 2 PO 4 , 10 MgSO 4 , 0.5 CaCl 2 and 10 glucose [12]. Slices were allowed to recover in standard ACSF (126 mM NaCl, 2.5 mM KCl, 1.25 mM NaH 2 PO 4 , 2 mM MgCl 2 , 2 mM CaCl 2 , 10 mM glucose and 26 mM NaHCO 3 ) oxygenated with 5% CO 2 and 95% O 2 for 1–2 h.
2.3. Fluorescence recovery after photobleaching ( FRAP) assay Slices prepared as described above were loaded with CDCF (dicarboxydichlorofluorescein diacetate), a gap junction- and cell-membrane-permeable dye that becomes membrane-impermeable after de-esterification. Brain slices were incubated in 1 M of the dye dissolved in ACSF. After several washes in the same solution in the absence of CDCF (to allow for complete de-esterification), the slices were placed in the stage of an inverted microscope attached to a Bio-Rad 1000 confocal microscope (Hercules, CA, USA). After excitation with the 488 nm line of the krypton / argon laser, a baseline fluorescence was collected using a 403 objective. Then, the area of laser scanning was reduced and the fluorescence signal bleached using 103 zoom and full power of laser. After three to five scans, the normal recording configuration was reestablished and refill was recorded for 40 s [11,15]. All studies were performed using optical sections 30 mm below the surface to ensure that the cells of interest had not been damaged during preparation. The FRAP technique does not discriminate between cell types. However, when utilized in adult brain tissue, fluorescence refill primarily reflects astrocytic coupling, as (1) only astrocytes are coupled to a significant extent in adult brain and (2) de-esterification of fluorescence indicators, including
A rabbit anti-mouse Cx43 (kindly provided by A. Lau), and a rabbit anti-rat Cx30 antibodies (Zymed, catalog No. 71-2200) were used. Eight mm cryostat sections of fresh brain tissue were fixed with methanol [15,14]. After overnight incubation with a 1:1000 dilution of anti-Cx43 antibody and a 1:250 dilution for anti-Cx30 antibody, detection was performed by applying a goat anti-rabbit secondary antibody conjugated to FITC (Jackson). Slices were mounted in Slow Fade (Molecular Probes) and examined by confocal microscopy (MRC1000; Bio-Rad) through a Nikon Diaphot inverted scope. Image analysis was performed using the Q500MC program from Leica. Analayiss of variance (ANOVA) was used to compare mean values whereas Student’s t-test was applied to determine which groups significantly differed. Number of Cx-immunoreactive plaques per 0.2 mm 2 was quantified in a minimum of 5 fields in each animal.
2.5. Western blot analysis Brain slices were homogenized by a tissue grinder in 10 mM Tris buffer and lysed in 2% sodium dodecyl sulphate (SDS) buffer [10]. After determination of the protein content by the Bradford assay (Bio-Rad) with bovine serum albumin (BSA) as standard, samples were diluted with 33 sodium dodecyl sulfate (SDS) gel-loading buffer. For detection of connexins, 50 mg of homogenate protein was resolved on 10% SDS–polyacrylamide gel electrophoresis (PAGE) gels, transferred to nitrocellulose membranes, blocked overnight at 48C in TBST (Tris, NaCl and Tween-20) with 3% BSA and incubated for 2 h at room temperature with the polyclonal anti-Cx43 1:1000) or the anti-Cx30 1:250) antibody. After repeated washes, an horesradish peroxidase (HRP)-conjugated goat anti-rabbit IgG was applied and antibody binding visualized by using an enhanced chemiluminescence kit (Amersham).
3. Results
3.1. Connexin-43 expression is higher in slices from senescent than from young hippocampus Immunohistochemical analysis of hippocampal sections from young adult mice (3 months), adult (7 months) and from old mouse brain (21 months) revealed that Cx43, the major gap junction protein of astrocytes, was organized in immunoreactive plaques as expected (Fig. 1). In accordance with earlier reports [16,15], the Cx-plaques were localized primarily to the plasma membrane, a pattern identical to the one observed in dissociated cultures of
M.L. Cotrina et al. / Brain Research 901 (2001) 55 – 61
57
Fig. 1. The expression of gap junction proteins increases as the brain matures. Immunohistochemistry of hippocampal sections from 3-, 7- and 21-month-old mice with antibodies directed against (A) Cx43 or (B) Cx30. Both the expression of Cx43 and Cx30 peaked in 7-month-old mice. Scale bar, 5 mm. (C) Bar histogram of the overall immunoreactivity of Cx43 and Cx30 in hippocampus of mice of the same ages as illustrated in A and B. Rel. immunoreactivity indicates mean number of plaques per 0.2 mm 2 . * Cx43, significantly different from 3-month-old mice. ** Cx30, significantly different from 3-month-old mice. P,0.05 by t-test assay.
cortical astrocytes [5]. Interestingly, higher levels of Cx43 immunoreactivity were evident in adult vs. young animals. Although the number of immunoreactive Cx43 plaques detected in the older animals was not significantly higher than in the younger animals (596661 plaques / 0.2 mm 2 in 3-month-old mice vs. 734662 / 0.2 mm 2 and 7556114 / 0.2 mm 2 plaques in 7- and 21-month-old mice, respectively; P.0.1 Student’s t-test), the size of the gap junction plaques increased as a function of age (0.960.1 and 1.360.1 mm 2 in 3- and 7-month-old mice, respectively;
P,0.02 Student’s t-test). However, the increase in the number of plaques was inversely proportional to the size of the plaques in the senescent brain (0.760.1 mm 2 , Fig. 1). A similar picture emerged when we analyzed the expression pattern of Cx30, which reaches maximal expression in the adult rat brain [14]. High levels of Cx30 were also detected in the brain of adult mice. In fact, a twofold increase in connexin expression was evident in adult compared to younger animals (315649 plaques / 0.2 mm 2 , size 0.860.07 mm 2 vs. 585651 / 0.2 mm 2 plaques,
58
M.L. Cotrina et al. / Brain Research 901 (2001) 55 – 61
size 0.960.1 mm 2 in 3- and 7-month-old mice, respectively; P,0.001 Student’s t-test) but only 262663 / 0.2 mm 2 plaques (size 0.460.04 mm 2 ) were detected in the brain of old mice. These data indicate that there is an overall loss of immunoreactivity against both Cx43 and Cx30 in older rats. Specifically, the size of the Cx43 immunoreactive plaques was reduced, whereas there was a decrease in both the number and size of the Cx30 plaques in old mice.
3.2. Gap junctions are functional in the aging brain Measurement of functional coupling in intact mouse brain slices using FRAP demonstrated that functional coupling was not significantly altered as a function of aging (Fig. 2). Average recovery of fluorescence after 2 min was 6366% in younger animals, (three mice, 3 months, a total of 13 FRAP experiments), 5965% in adult animals (two mice, 7 months, a total of 16 FRAP experiments) and 5464% in old brain (two mice, 19 and 21 months, a total of 15 FRAP experiments). Although gap junction permeability did not significantly change, an overall tendency to decreased coupling was observed as the animal aged.
3.3. Western blot analysis of Cx43 and Cx30 The results presented above document that the number and / or size of Cx43 and Cx30 immunoreactive plaques decline during the aging process in hippocampus. We next tested whether the decline in Cx-plaques correlated with the amount of connexin protein present in brain. To this end, we used Western blot analysis of whole brain extracts. Fig. 3 shows that the levels of both Cx43 and Cx30 remained relatively constant throughout the lifespan of the mice studied. Thus, the decline in Cx43 and Cx30 immunoreactive plaques does not result from a general decrease in the pool of Cx-proteins, but rather from redistribution. Of note, in accordance with earlier reports, we only observed the unphosphorylated 41 kDa band and not the 43 kDa phosphorylated band [10]. Brain tissue contains endogenous phosphatases that rapidly unphosphorylate Cx43 during tissue preparation, unless brain metabolism is rapidly inactivated by cranial high-energy microwave irradiation or inclusion of phosphatase inhibitors [10].
4. Discussion Little information exists with regard to the aging process of astrocytes and no studies have explored so far how well these cells communicate in the senescent brain. Here we have mapped the expression levels of the most abundant astrocytic gap junction proteins, Cx43 and Cx30, and analyzed the extent of functional coupling in young versus
aged brain tissue. Our data show that the levels of Cx43, a gap junction protein that is expressed from very early in development in most astrocytes [16], decreased in aging. Aging was associated with a reduction in the average size of Cx43 immunoreactive plaques, whereas the number of plaques remained essentially unchanged in senescent brains (21-month-old mice). A slightly different pattern of age-induced modulation of another member of the Cxfamily, Cx30, was observed. Cx30, a gap junction protein with a late onset of expression in the adult murine brain [14], is indeed expressed in young adult mice (3-monthold). Its expression is increased in adult mice, but decreases again as the brain ages. In fact, both the number of Cx30-immunoreactive plaques, as well as the size of the plaques were clearly reduced in old tissue. In rat hippocampus, aging has been associated with an increase in the number and size of astrocytes [2]. It is therefore not possible to estimate the relative number of Cx-plaques per astrocytes as a function of aging. We used the FRAP technique to evaluate functional gap junction coupling in freshly prepared hippocampal slices. The logic for using this approach was that intracellular injection of gap-junction-permeable tracers becomes increasingly difficult as a function of the age of the mice brain (our unpublished observations). We were concerned that the procedures associated with tracer injection would affect the outcome of the analysis and preferentially would decrease tracer spread in the aging brain. We therefore chose to analyze gap junction coupling using the noninvasive FRAP technique. The fact that coupling remained relatively high in the older animals suggests that our measurements accurately reflect that astrocytes remain coupled during the aging process. It is interesting to note that the size of the gap junction plaques was reduced in aged brain. The reduction in plaque size was not associated with a significant decrease in functional coupling, suggesting that the reduction in plaque size did not significantly impair channel function. However, as the brain ages, a reduction in plaque size may be indicative of a modulation, rather than a loss of connexin function. A similar dissociation between reduction of connexin levels but constant intercellular communication has been recently observed in mammary epithelial cells undergoing changes in their activation state [18]. In this regard, alternative roles for connexins that do not require the formation of gap junction channels have been described in rat astrocytic cultures [6,8]. In ischemic brain tissue, aggregation, rather than disassembly of gap junction plaques was observed in astrocytes during the process of cell death [5]. In this study, the dramatic increase in plaque size was associated with a clear reduction in the degree of phosphorylated Cx43 and a consequent loss of channel permeability. Alternatively, the modulation we observe in the number and size of the gap junction plaques might represent a reorganization of the plaques within the network of autocellular contacts of the astrocytes. Indeed,
M.L. Cotrina et al. / Brain Research 901 (2001) 55 – 61
59
Fig. 2. Gap junction functionality is preserved in brains from old mice. The fluorescence recovery after photobleach technique (FRAP) was used to quantify gap junction function. Top panels illustrate hippocampal slices from 3-month-old (left) and 21-month-old mice (right) loaded with the gap junction permeable tracer CDCF. Middle panel illustrates the area selected for laser bleaching (red rectangle). Lower panels show refill of fluorescence 2 min after laser bleach (arrow). Bar histogram maps the percentage of recovery after photobleach in 3-, 7- and 21-month-old mice. Fluorescence recovery does not differ significantly between the different ages, although a trend towards a decrease is evident in older mice.
60
M.L. Cotrina et al. / Brain Research 901 (2001) 55 – 61
References
Fig. 3. Western blot analysis of gap junction expression in aging. Lane 1, 3-month-old mice; lane 2, 7-month-old mice; lane 3, 21-month-old mice. The expression levels of both Cx43 and Cx30 remained relatively constant.
reflexive gap junctions have been reported to represent a considerable amount of the total astrocytic gap junctions [21]. Although the meaning of reflexive gap junctions is still unknown, changes in the organization of these connections would not necessarily affect gap junction permeability. In conclusion, although astrocytic gap junction plaques are reduced in the brains of old mice compared to young adults, functional coupling remains high. The reorganization of Cx-immunoreactive plaques in the senescent brain, which is expressed as a redistribution in plaque size (Cx43) or a reduction of immunoreactive plaques (Cx30), suggests that functions other than channel permeability are modulated in aging. The observation provides an experimental framework for the manipulation of gap junction-dependent glial calcium signaling and, by extension, other connexin-dependent neural processes that may be impaired in aging.
Acknowledgements This study was supported by NINDS / NIH grants NS30007 and NS35011. M.N. is an Established Investigator sponsored by The American Heart Association.
[1] V. Alvarez-Maubecin, F. Garcia-Hernandez, J.T. Williams, E.J. Van Bockstaele, Functional coupling between neurons and glia, J. Neurosci. 20 (2000) 4091–4098. [2] F. Amenta, E. Bronzetti, M. Sabbatini, J.A. Vega, Astrocyte changes in aging cerebral cortex and hippocampus: a quantitative immunohistochemical study, Microsc. Res. Tech. 43 (1998) 29–33. [3] A. Araque, V. Parpura, R.P. Sanzgini, P.G. Haydon, Tripartite synapses: glia, the unacknowledged partner, Trends Neurosci. 22 (1999) 208–215. [4] G. Arcuino, S.A. Goldman, M.L. Cotrina, S. Liu, V. Fried, J.H.-L. Lin, M. Nedergaard, Connexin can direct selective cell–cell interactions, Soc. Neurosi. Abstr. 25 (1999) 600.11. [5] M.L. Cotrina, J. Kang, J.H.-C. Lin, E. Bueno, T.W. Hansen, L. He, Y. Liu, M. Nedergaard, Astrocytic gap junctions remain open during ischemic conditions, J. Neurosci. 18 (1998) 2520–2537. [6] M.L. Cotrina, J.H.-C. Lin, M. Nedergaard, Cytoskeletal assembly and ATP release regulate astrocytic calcium signaling, J. Neurosci. 18 (1998) 8794–8804. [7] M.L. Cotrina, J.H.-C. Lin, A. Alves-Rodrigues, S. Liu, J. Li, H. Azmi-Ghadimi, J. Kang, C.C.G. Naus, M. Nedergaard, Connexins regulate calcium signaling by controlling ATP release, Proc. Natl. Acad. Sci. USA 95 (1998) 15735–15740. ´ [8] M.L. Cotrina, J.H.-C. Lin, J.C. Lopez-Garcia, C.C.G. Naus, M. Nedergaard, Connexin-mediated glia signaling, J. Neurosci. 20 (2000) 2835–2844. [9] R. Dermietzel, D. Spray, From neuro-glue (‘nervenkitt’) to glia: a prologue, Glia 24 (1998) 1–7. [10] M.Z. Hossain, L.J. Murphy, E.L. Hertzberg, J.I. Nagy, Phosphorylated forms of connexin 43 predominate in rat brain: demonstration by rapid inactivation of brain metabolism, J. Neurochem. 62 (1994) 2394–2403. [11] G.Y. Huang, E.S. Cooper, K. Waldo, M.L. Kirby, N.B. Gilula, C.W. Lo, Gap junction mediated cell–cell communication modulates mouse neural crest migration, J. Cell Biol. 143 (1998) 1725–1734. [12] J. Kang, L. Jiang, S.A. Goldman, M. Nedergaard, Astrocyte-mediated potentiation of inhibitory synaptic transmission, Nat. Neurosci. 1 (1998) 683–692. [13] J. Kang, M. Nedergaard, Imaging astrocytes in acute brain slices, in: R. Yuste, F. Lanni, A. Konnerth (Eds.), Imaging Structure and Function in the Nervous System, Cold Spring Harbor Laboratory Press, 2000, pp. 42.1–42.11. [14] P. Kunzelmann, W. Schroder, O. Traub, C. Steinhauser, R. Dermietzel, K. Willecke, Late onset and increasing expression of the gap junction protein connexin30 in adult murine brain and long-term cultured astrocytes, Glia 25 (1999) 111–119. [15] J.H.-C. Lin, H. Weigel, M.L. Cotrina, S. Liu, E. Bueno, A.J. Hansen, T.W. Hansen, S. Goldman, M. Nedergaard, Gap-junction-mediated propagation and amplification of cell injury, Nat. Neurosci. 6 (1998) 494–500. [16] B. Nadarajah, D. Thomaidou, W.H. Evans, J.G. Parnavelas, Gap junctions in the adult cerebral cortex: regional differences in their distribution and cellular expression of connexins, J. Comp. Neurol. 376 (1996) 326–342. [17] B.R. Ransom, Do glial gap junctions play a role in extracellular ion homeostasis, in: D.C. Spray, R. Dermietzel (Eds.), Gap Junctions in the Nervous System, R.R. Landes, Austin, TX, 1996, pp. 160–173. [18] M.A. Sia, T.L. Woodward, J.D. Turner, D.W. Laird, Quiescent mammary epithelial cells have reduced Cx43 but maintain a high level of gap junction intercellular communication, Dev. Genet. 24 (1999) 111–122. [19] S.J. Smith, Neuromodulatory astrocytes, Curr. Biol. 4 (1994) 807– 810. [20] A. Tabernero, C. Giaume, J.M. Medina, Endothelin-1 regulates glucose utilization in cultured astrocytes by controlling intercellular communication through gap junctions, Glia 16 (1996) 187–195.
M.L. Cotrina et al. / Brain Research 901 (2001) 55 – 61 [21] J.R. Wolff, K. Stuke, M. Missler, H. Tytko, P. Schwarz, A. Rohlmann, T.I. Chao, Autocellular coupling by gap junctions in cultured astrocytes: a new view on cellular autoregulation during process formation, Glia 24 (1998) 121–140.
61
[22] T. Yamamoto, E. Ochalski, E. Hertzberg, J. Nagy, On the organization of astrocytic gap junctions in rat brain as suggested by LM and EM immunohistochemistry of connexin43 expression, J. Comp. Neurol. 302 (1990) 853–883.