- Email: [email protected]
Changes in cortical noradrenergic axon terminals of locus coeruleus neurons in aged F344 rats
Neuroscience Letters 307 (2001) 197±199
www.elsevier.com/locate/neulet
Changes in cortical noradrenergic axon terminals of locus coeruleus neurons in aged F344 rats Y. Ishida a,1, T. Shirokawa a,*, Y. Komatsu b, K. Isobe a a
Laboratory of Physiology, Department of Basic Gerontology, National Institute for Longevity Sciences, Gengo 36-3, Morioka-cho, Obu 474-8522, Japan b Department of Visual Neuroscience, Division of Higher Order Nervous Function Research, Institute of Environmental Medicine, Nagoya University, Nagoya, Japan Received 7 April 2001; received in revised form 23 May 2001; accepted 24 May 2001
Abstract The noradrenergic innervations and noradrenaline contents of the frontal cortex in two age groups (9 and 25 months) of male F344 rats have been quanti®ed by electrophysiological and biochemical methods. In the electrophysiological study, the percentage of locus coeruleus (LC) neurons activated antidromically from the frontal cortex decreased with age. In contrast, the percentage of LC neurons showing multiple antidromic latencies, which suggests axonal branching of individual LC neurons, increased markedly between 9 and 25 months in the frontal cortex. In the biochemical study, we found no signi®cant difference in noradrenaline levels in the cortical terminal ®elds of LC neurons during aging. These results suggest that LC neurons give rise to axonal branches to retain noradrenaline levels in their target ®elds in the aged brain. Our ®ndings show that LC neurons preserve a strong capability for remodeling their axon terminals even in the aged brain. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Aging; Locus coeruleus; Noradrenergic axon; Noradrenaline; F344 rat
The locus coeruleus (LC) is the largest noradrenergic nucleus in the central nervous system [1]. It is known as a major target region such as the frontal cortex. Although the function of noradrenergic input from the LC to the frontal cortex is still unclear, the role of LC has been suggested to involve integrative neuronal functions such as vigilance [2]. These functions decline during the normal aging process. Recently, we have obtained evidence suggesting that LC neurons might give rise to axonal branching following the loss of projection to the frontal cortex [5,10]. However, it remains unclear whether such electrophysiological changes are accompanied by changes in noradrenaline levels in the frontal cortex. To investigate the age-dependent changes in the projections from the LC to the frontal cortex, we employed in vivo electrophysiological techniques to antidromically activate the cortical axon terminals. In addition, noradrenaline contents in the cortical target ®elds (the fron* Corresponding author. Tel.: 181-562-46-2311; fax: 181-56248-2373. E-mail address: [email protected] (T. Shirokawa). 1 Present address: Department of Physiology, Yamaguchi University School of Medicine, Ube, Yamaguchi 755-8505, Japan.
tal cortex and visual cortex) were assayed by high performance liquid chromatography (HPLC) with electrochemical detection. Male F344/N rats (two groups; 9 and 25 months of age, 12 animals in each age group) were housed with food and water available ad libitum on a 12-h light/dark cycle. All animal procedures complied with the guidelines of the Animal Research Facilities Committee of the National Institute for Longevity Sciences. For the electrophysiological assays, six animals from each age group were anesthetized with urethane (1.2 g/kg, i.p.). Lidocaine (4% Xylocaine) was applied locally to all incisions. Rectal temperature was maintained at 36.58C by a heating pad. Electrocardiogram (ECG) and electroencephalogram (EEG) were monitored continuously during the experiments. Stimulating electrodes of the bipolar type consisted of two insulated stainless steel wires (diameter 200 mm, tip separation 0.5 mm). Electrical pulses were 0.5 ms in duration with currents ranging from 0.1 to 6.0 mA, and the cycle of stimulation was 1.5 s. The electrodes were stereotaxically guided into the frontal cortex (Fr2; AP 3.0 mm, L 1.5 mm, D 1.2 mm) [9]. The electrical activity of LC neurons was extracellularly recorded with glass pipette microelectrodes ®lled with 2 M
0304-3940/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 1) 01 96 3- 2
198
Y. Ishida et al. / Neuroscience Letters 307 (2001) 197±199
NaCl. Electrode resistance ranged from 10 to 20 MV. LC neurons were usually encountered 6.0±6.5 mm below the cortical surface. In each animal, 60±66 LC neurons were recorded by moving a recording electrode within about 100 mm rostrocaudally or mediolaterally to avoid sampling bias. The LC neurons were identi®ed according to the criteria [5±7,10,11]. Brie¯y, the LC neurons revealed a wide spike duration (~ 2 ms), a slow and tonic rate of spontaneous ®ring (0.5±6 spikes/s), and excitation by tail pinches followed by a long-lasting suppression of ®ring. Responses of LC neurons were assumed to be antidromic provided that the following criteria were satis®ed: (1) ®xed latency; (2) ability to follow high-frequency stimulation (. 200 Hz), and, most importantly; (3) collision with spontaneous spikes [5±7,10,11]. We employed two electrophysiological measurements: ®rst; to quantify the density of LC axons projecting to the frontal cortex, the percentage of LC neurons activated antidromically from the frontal cortex was calculated (Pindex number of LC neurons with antidromic latencies / number of recorded LC neurons). Second, to quantify the degree of branching of LC axon terminals in the frontal cortex, we calculated the ratio of LC neurons showing two or more discrete antidromic latencies (which suggests axonal branching of individual LC neurons) at different intensities of stimulus currents (multiple antidromic latencies [7]) (M-index number of LC neurons with multiple
antidromic latencies / number of LC neurons with antidromic latencies) [5,10]. For the HPLC assays, the brains of six animals from each age group were rapidly removed, and the frontal cortex, visual cortex and pons were dissected for biochemical analysis. Each sample was homogenized in a tube containing 0.2 M perchloric acid with an appropriate amount of internal standard (isoproterenol), and was spun at 16; 000 £ g for 30 min. After centrifugation, the supernatants were further ®ltered out (0.22 mm) and stored at 48C pending chemical analysis. Noradrenaline was assayed by HPLC with electrochemical detection (EICOM, Japan). Separation of NA was achieved on an EICOMPAK SC-50DS column (150 £ 3:0 mm i.d.) and a WE-GC carbon electrode set at 1750 mV vs. an Ag/ AgCl reference electrode. The column temperature was kept at 258C. The mobile phase was 0.1 M citric acid ± 0.1 M sodium acetate (pH 3.5) containing 17% methanol, 100 mg/ml 1-octanesulfonic acid, and 5 mg/ml EDTA (2Na). The data were expressed as means ^ SEM, and were compared by Student's t-test. The total number of LC neurons recorded from the six animals in the 9 and 25 month age groups, respectively, were; n 380 and n 379. Of the total LC neurons recorded, the number activated antidromically from the frontal cortex for those respective groups was; 9 months, n 225; 25 months, n 130. The number of LC neurons with multiple antidromic latencies for those groups was; 9
Fig. 1. (A) The P-index in the groups of 9-month- (9m) and 25-month- (25m)-old rats. Horizontal bar indicates the mean P-index of six animals from each age group. The mean P-index in the 25-m rats was signi®cantly lower than that obtained in the 9-m group (**P , 0:01). (B) The M-index in each age group. Horizontal bar indicates the mean M-index of six animals from each age group. The mean M-index in the 25-m group was signi®cantly higher than that obtained in the 9-m group (**P , 0:01).
Y. Ishida et al. / Neuroscience Letters 307 (2001) 197±199
199
®elds of LC neurons did not change with age. One likely interpretation of these results is that the increased axonal branches compensate for the loss of LC projections so as to retain noradrenaline levels in the terminal ®elds. An alternative interpretation is that the excitability of axon terminals of LC neurons increased in the frontal cortex. In fact, we observed just such an increase in the aged rat [11]. Since the excitability of LC axon terminals is probably linked to the transmitter release [8], it is likely that the increased terminal excitability causes an increase in noradrenaline release in the aged brain. Another possibility is that the compensatory upregulation of noradrenaline synthesis increases with age. However, this is unlikely because mRNA levels of tyrosine hydroxylase (a catecholamine-synthesizing enzyme) in the LC are not signi®cantly different between young and aged F344 rats [3]. Taken together, these ®ndings suggest that the LC neurons preserve a strong capability for remodeling their axon terminals even in the aged brain.
Fig. 2. Effect of age on noradrenaline contents in the frontal cortex, visual cortex and pons. Data are expressed as ng/g wet weight ^ SEM (n 6). There was no signi®cant difference in the noradrenaline levels in each brain region between age groups.
months, n 87; 25 months, n 91. The P-index revealed a marked decrease between 9 and 25 months of age (Fig. 1A). The P-index obtained in the 25-month group (35.6 ^ 1.9, n 6) was signi®cantly lower than in the 9-month group (59.3 ^ 3.8, n 6) (P , 0:01). In contrast, the M-index obtained in the 25-month group (70.2 ^ 4.3, n 6) was signi®cantly higher than that obtained in the 9-month group (36.3 ^ 4.3, n 6) (P , 0:01) (Fig. 1B). The HPLC assays revealed a difference in noradrenaline contents among the frontal cortex, visual cortex and pons during aging. However, there was no signi®cant difference between age groups in the noradrenaline levels for each brain region (Fig. 2). In the present study, the electrophysiological data clearly show that the projections from LC to the frontal cortex decrease with age, whereas a high degree of arborization occurs in the axon terminals of individual LC neurons. Recently, we found that the density of noradrenergic axons and varicosities (swellings along an axon from which noradrenaline is released) decreases with age in the frontal cortex without changes in the number of varicosities per unit axon length [4]. This age-dependent decrease in projections from LC to the frontal cortex is followed by an increase in axonal branching after 17 months of age [5,10]. Our present electrophysiological data agree with these studies. On the other hand, the biochemical data show that the noradrenaline contents in the cortical target
[1] Amaral, D.G. and Sinnamon, H.E., The locus coeruleus: neurobiology of a central noradrenergic nucleus, Prog. Neurobiol., 9 (1977) 147±196. [2] Aston-Jones, G. and Bloom, F., Norepinephrine-containing locus coeruleus neurons in behaving rats exhibit pronounced responses to non-noxious environmental stimuli, Neuroscience, 1 (1981) 887±900. [3] Cizza, G., Pacak, K., Kvetnansky, R., Palkovits, M., Goldstein, D.S., Brady, L.S., Fukuhara, K., Bergamini, E., Kopin, I.J., Blackman, M.R., Chrousos, G.P. and Gold, P.W., Decreased stress responsivity of central and peripheral catecholaminergic systems in aged 344/N Fischer rats, J. Clin. Invest., 95 (1995) 1217±1224. [4] Ishida, Y., Shirokawa, T., Miyaishi, O., Komatsu, Y. and Isobe, K., Age-dependent changes in noradrenergic innervations of the frontal cortex in F344 rats, Neurobiol. Aging, 22 (2001) 283±286. [5] Ishida, Y., Shirokawa, T., Miyaishi, O., Komatsu, Y. and Isobe, K., Age-dependent changes in projections from locus coeruleus to hippocampus dentate gyrus and frontal cortex, Eur. J. Neurosci., 12 (2000) 1263±1270. [6] Nakamura, S., Some electrophysiological properties of neurons of rat locus coeruleus, J. Physiol., 267 (1977) 641±658. [7] Nakamura, S., Sakaguchi, T. and Aoki, F., Electrophysiological evidence for terminal sprouting of locus coeruleus neurons following repeated mild stress, Neurosci. Lett., 100 (1989) 147±152. [8] Nakamura, S., Tepper, J.M., Young, S.J. and Groves, P.M., Neurophysiological consequences of presynaptic receptor activation: changes in noradrenergic terminal excitability, Brain Res., 226 (1981) 155±170. [9] Paxinos, G. and Watson, C., The rat brain in stereotaxic coordinates, , 2nd EditionAcademic Press, San Diego, 1986. [10] Shirokawa, T., Ishida, Y. and Isobe, K., Age-dependent changes in axonal branching of single locus coeruleus neurons projecting to two different terminal ®elds, J. Neurophysiol., 84 (2000) 1120±1122. [11] Shirokawa, T., Ishida, Y. and Isobe, K., Changes in electrophysiological properties of axon terminals of locus coeruleus neurons with age in F344 rat, Neurosci. Lett., 289 (2000) 69±71.
www.elsevier.com/locate/neulet
Changes in cortical noradrenergic axon terminals of locus coeruleus neurons in aged F344 rats Y. Ishida a,1, T. Shirokawa a,*, Y. Komatsu b, K. Isobe a a
Laboratory of Physiology, Department of Basic Gerontology, National Institute for Longevity Sciences, Gengo 36-3, Morioka-cho, Obu 474-8522, Japan b Department of Visual Neuroscience, Division of Higher Order Nervous Function Research, Institute of Environmental Medicine, Nagoya University, Nagoya, Japan Received 7 April 2001; received in revised form 23 May 2001; accepted 24 May 2001
Abstract The noradrenergic innervations and noradrenaline contents of the frontal cortex in two age groups (9 and 25 months) of male F344 rats have been quanti®ed by electrophysiological and biochemical methods. In the electrophysiological study, the percentage of locus coeruleus (LC) neurons activated antidromically from the frontal cortex decreased with age. In contrast, the percentage of LC neurons showing multiple antidromic latencies, which suggests axonal branching of individual LC neurons, increased markedly between 9 and 25 months in the frontal cortex. In the biochemical study, we found no signi®cant difference in noradrenaline levels in the cortical terminal ®elds of LC neurons during aging. These results suggest that LC neurons give rise to axonal branches to retain noradrenaline levels in their target ®elds in the aged brain. Our ®ndings show that LC neurons preserve a strong capability for remodeling their axon terminals even in the aged brain. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Aging; Locus coeruleus; Noradrenergic axon; Noradrenaline; F344 rat
The locus coeruleus (LC) is the largest noradrenergic nucleus in the central nervous system [1]. It is known as a major target region such as the frontal cortex. Although the function of noradrenergic input from the LC to the frontal cortex is still unclear, the role of LC has been suggested to involve integrative neuronal functions such as vigilance [2]. These functions decline during the normal aging process. Recently, we have obtained evidence suggesting that LC neurons might give rise to axonal branching following the loss of projection to the frontal cortex [5,10]. However, it remains unclear whether such electrophysiological changes are accompanied by changes in noradrenaline levels in the frontal cortex. To investigate the age-dependent changes in the projections from the LC to the frontal cortex, we employed in vivo electrophysiological techniques to antidromically activate the cortical axon terminals. In addition, noradrenaline contents in the cortical target ®elds (the fron* Corresponding author. Tel.: 181-562-46-2311; fax: 181-56248-2373. E-mail address: [email protected] (T. Shirokawa). 1 Present address: Department of Physiology, Yamaguchi University School of Medicine, Ube, Yamaguchi 755-8505, Japan.
tal cortex and visual cortex) were assayed by high performance liquid chromatography (HPLC) with electrochemical detection. Male F344/N rats (two groups; 9 and 25 months of age, 12 animals in each age group) were housed with food and water available ad libitum on a 12-h light/dark cycle. All animal procedures complied with the guidelines of the Animal Research Facilities Committee of the National Institute for Longevity Sciences. For the electrophysiological assays, six animals from each age group were anesthetized with urethane (1.2 g/kg, i.p.). Lidocaine (4% Xylocaine) was applied locally to all incisions. Rectal temperature was maintained at 36.58C by a heating pad. Electrocardiogram (ECG) and electroencephalogram (EEG) were monitored continuously during the experiments. Stimulating electrodes of the bipolar type consisted of two insulated stainless steel wires (diameter 200 mm, tip separation 0.5 mm). Electrical pulses were 0.5 ms in duration with currents ranging from 0.1 to 6.0 mA, and the cycle of stimulation was 1.5 s. The electrodes were stereotaxically guided into the frontal cortex (Fr2; AP 3.0 mm, L 1.5 mm, D 1.2 mm) [9]. The electrical activity of LC neurons was extracellularly recorded with glass pipette microelectrodes ®lled with 2 M
0304-3940/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 1) 01 96 3- 2
198
Y. Ishida et al. / Neuroscience Letters 307 (2001) 197±199
NaCl. Electrode resistance ranged from 10 to 20 MV. LC neurons were usually encountered 6.0±6.5 mm below the cortical surface. In each animal, 60±66 LC neurons were recorded by moving a recording electrode within about 100 mm rostrocaudally or mediolaterally to avoid sampling bias. The LC neurons were identi®ed according to the criteria [5±7,10,11]. Brie¯y, the LC neurons revealed a wide spike duration (~ 2 ms), a slow and tonic rate of spontaneous ®ring (0.5±6 spikes/s), and excitation by tail pinches followed by a long-lasting suppression of ®ring. Responses of LC neurons were assumed to be antidromic provided that the following criteria were satis®ed: (1) ®xed latency; (2) ability to follow high-frequency stimulation (. 200 Hz), and, most importantly; (3) collision with spontaneous spikes [5±7,10,11]. We employed two electrophysiological measurements: ®rst; to quantify the density of LC axons projecting to the frontal cortex, the percentage of LC neurons activated antidromically from the frontal cortex was calculated (Pindex number of LC neurons with antidromic latencies / number of recorded LC neurons). Second, to quantify the degree of branching of LC axon terminals in the frontal cortex, we calculated the ratio of LC neurons showing two or more discrete antidromic latencies (which suggests axonal branching of individual LC neurons) at different intensities of stimulus currents (multiple antidromic latencies [7]) (M-index number of LC neurons with multiple
antidromic latencies / number of LC neurons with antidromic latencies) [5,10]. For the HPLC assays, the brains of six animals from each age group were rapidly removed, and the frontal cortex, visual cortex and pons were dissected for biochemical analysis. Each sample was homogenized in a tube containing 0.2 M perchloric acid with an appropriate amount of internal standard (isoproterenol), and was spun at 16; 000 £ g for 30 min. After centrifugation, the supernatants were further ®ltered out (0.22 mm) and stored at 48C pending chemical analysis. Noradrenaline was assayed by HPLC with electrochemical detection (EICOM, Japan). Separation of NA was achieved on an EICOMPAK SC-50DS column (150 £ 3:0 mm i.d.) and a WE-GC carbon electrode set at 1750 mV vs. an Ag/ AgCl reference electrode. The column temperature was kept at 258C. The mobile phase was 0.1 M citric acid ± 0.1 M sodium acetate (pH 3.5) containing 17% methanol, 100 mg/ml 1-octanesulfonic acid, and 5 mg/ml EDTA (2Na). The data were expressed as means ^ SEM, and were compared by Student's t-test. The total number of LC neurons recorded from the six animals in the 9 and 25 month age groups, respectively, were; n 380 and n 379. Of the total LC neurons recorded, the number activated antidromically from the frontal cortex for those respective groups was; 9 months, n 225; 25 months, n 130. The number of LC neurons with multiple antidromic latencies for those groups was; 9
Fig. 1. (A) The P-index in the groups of 9-month- (9m) and 25-month- (25m)-old rats. Horizontal bar indicates the mean P-index of six animals from each age group. The mean P-index in the 25-m rats was signi®cantly lower than that obtained in the 9-m group (**P , 0:01). (B) The M-index in each age group. Horizontal bar indicates the mean M-index of six animals from each age group. The mean M-index in the 25-m group was signi®cantly higher than that obtained in the 9-m group (**P , 0:01).
Y. Ishida et al. / Neuroscience Letters 307 (2001) 197±199
199
®elds of LC neurons did not change with age. One likely interpretation of these results is that the increased axonal branches compensate for the loss of LC projections so as to retain noradrenaline levels in the terminal ®elds. An alternative interpretation is that the excitability of axon terminals of LC neurons increased in the frontal cortex. In fact, we observed just such an increase in the aged rat [11]. Since the excitability of LC axon terminals is probably linked to the transmitter release [8], it is likely that the increased terminal excitability causes an increase in noradrenaline release in the aged brain. Another possibility is that the compensatory upregulation of noradrenaline synthesis increases with age. However, this is unlikely because mRNA levels of tyrosine hydroxylase (a catecholamine-synthesizing enzyme) in the LC are not signi®cantly different between young and aged F344 rats [3]. Taken together, these ®ndings suggest that the LC neurons preserve a strong capability for remodeling their axon terminals even in the aged brain.
Fig. 2. Effect of age on noradrenaline contents in the frontal cortex, visual cortex and pons. Data are expressed as ng/g wet weight ^ SEM (n 6). There was no signi®cant difference in the noradrenaline levels in each brain region between age groups.
months, n 87; 25 months, n 91. The P-index revealed a marked decrease between 9 and 25 months of age (Fig. 1A). The P-index obtained in the 25-month group (35.6 ^ 1.9, n 6) was signi®cantly lower than in the 9-month group (59.3 ^ 3.8, n 6) (P , 0:01). In contrast, the M-index obtained in the 25-month group (70.2 ^ 4.3, n 6) was signi®cantly higher than that obtained in the 9-month group (36.3 ^ 4.3, n 6) (P , 0:01) (Fig. 1B). The HPLC assays revealed a difference in noradrenaline contents among the frontal cortex, visual cortex and pons during aging. However, there was no signi®cant difference between age groups in the noradrenaline levels for each brain region (Fig. 2). In the present study, the electrophysiological data clearly show that the projections from LC to the frontal cortex decrease with age, whereas a high degree of arborization occurs in the axon terminals of individual LC neurons. Recently, we found that the density of noradrenergic axons and varicosities (swellings along an axon from which noradrenaline is released) decreases with age in the frontal cortex without changes in the number of varicosities per unit axon length [4]. This age-dependent decrease in projections from LC to the frontal cortex is followed by an increase in axonal branching after 17 months of age [5,10]. Our present electrophysiological data agree with these studies. On the other hand, the biochemical data show that the noradrenaline contents in the cortical target
[1] Amaral, D.G. and Sinnamon, H.E., The locus coeruleus: neurobiology of a central noradrenergic nucleus, Prog. Neurobiol., 9 (1977) 147±196. [2] Aston-Jones, G. and Bloom, F., Norepinephrine-containing locus coeruleus neurons in behaving rats exhibit pronounced responses to non-noxious environmental stimuli, Neuroscience, 1 (1981) 887±900. [3] Cizza, G., Pacak, K., Kvetnansky, R., Palkovits, M., Goldstein, D.S., Brady, L.S., Fukuhara, K., Bergamini, E., Kopin, I.J., Blackman, M.R., Chrousos, G.P. and Gold, P.W., Decreased stress responsivity of central and peripheral catecholaminergic systems in aged 344/N Fischer rats, J. Clin. Invest., 95 (1995) 1217±1224. [4] Ishida, Y., Shirokawa, T., Miyaishi, O., Komatsu, Y. and Isobe, K., Age-dependent changes in noradrenergic innervations of the frontal cortex in F344 rats, Neurobiol. Aging, 22 (2001) 283±286. [5] Ishida, Y., Shirokawa, T., Miyaishi, O., Komatsu, Y. and Isobe, K., Age-dependent changes in projections from locus coeruleus to hippocampus dentate gyrus and frontal cortex, Eur. J. Neurosci., 12 (2000) 1263±1270. [6] Nakamura, S., Some electrophysiological properties of neurons of rat locus coeruleus, J. Physiol., 267 (1977) 641±658. [7] Nakamura, S., Sakaguchi, T. and Aoki, F., Electrophysiological evidence for terminal sprouting of locus coeruleus neurons following repeated mild stress, Neurosci. Lett., 100 (1989) 147±152. [8] Nakamura, S., Tepper, J.M., Young, S.J. and Groves, P.M., Neurophysiological consequences of presynaptic receptor activation: changes in noradrenergic terminal excitability, Brain Res., 226 (1981) 155±170. [9] Paxinos, G. and Watson, C., The rat brain in stereotaxic coordinates, , 2nd EditionAcademic Press, San Diego, 1986. [10] Shirokawa, T., Ishida, Y. and Isobe, K., Age-dependent changes in axonal branching of single locus coeruleus neurons projecting to two different terminal ®elds, J. Neurophysiol., 84 (2000) 1120±1122. [11] Shirokawa, T., Ishida, Y. and Isobe, K., Changes in electrophysiological properties of axon terminals of locus coeruleus neurons with age in F344 rat, Neurosci. Lett., 289 (2000) 69±71.