BRAIN RESEARCH ELSEVIER
Brain Research 705 (1995) 45-52
Research report
Seasonal- and temperature-dependent variation in CNS ascorbate and glutathione levels in anoxia-tolerant turtles Miguel A. P6rez-Pinz6n 1 Margaret E. Rice * Departments of Physiology and Neuroscience and Neurosurgery, NYU Medical Center, 550 First Ave., New York, NY 10016, USA Accepted 22 August 1995
Abstract We determined the ascorbic acid (ascorbate) and glutathione (GSH) contents of eight regions of the CNS from anoxia-tolerant turtles collected in summer and in winter. Ascorbate was of special interest because it is found in exceptionally high levels in the turtle CNS. The temperature-dependence of CNS ascorbate content was established by comparing levels in animals collected from two geographic zones with different average winter temperatures and in animals re-acclimated to different temperatures in the laboratory. The analytical method was liquid chromatography with electrochemical detection. Turtle ascorbate levels were 30-40% lower in animals acclimatized to winter (2°C) than to summer (23°C) in all regions of the CNS. Similarly, GSH levels were 20-30% lower in winter than in summer. Winter ascorbate levels were higher in turtles from Louisiana (19°C) than in turtles acclimatized to winter in Wisconsin (2°C). Summer and winter levels of ascorbate could be reversed by re-acclimating animals to cold (I°C) or warm (23°C) temperatures for at least one week. CNS water content did not differ between cold- and warm-acclimated turtles. Taken together, the data indicated that ascorbate and GSH undergo significant seasonal variation and that the catalyst for change is environmental temperature. Steady-state ascorbate content showed a linear dependence on temperature, with a slope of 1.5% per °C that was independent of CNS region. Lower levels of cerebral antioxidants in turtles exposed to colder temperatures were consistent with the decreased rate of cerebral metabolism that accompanies winter hibernation. Cerebral ascorbate and GSH levels in the turtle remained similar to or higher than those in mammals, even during winter, however. These findings support the notion that unique mechanisms of antioxidant regulation in the turtle contribute to their tolerance of the hypoxia-reoxygenation that characterizes diving behavior. Keywords." Ascorbic acid; Free radical; Glutathione; Ischemia; Turtle; Rat; Temperature; Metabolism
1. Introduction Pond turtles have a remarkable tolerance of hypoxia [2,21,26,27,44]. A variety of mechanisms are thought to contribute to this process. In turtle brain tissue, these include lower basal rates of oxidative metabolism and glucose utilization compared to rat [54,59], as well as compensatory mechanisms that accompany hypoxia, such as increases in glycolytic rate, decreases in electrical activity and Na+-channel density, and the release of inhibitory neurotransmitters (for reviews see [18,26,27,50]). These processes work together to prevent loss of cellular homeostasis during hypoxia. By contrast, when mammalian
* Corresponding author. Fax: (1) (212) 689-0334; e-mail:
[email protected]. 1 Present address: Dept. of Neurology, D4-5, University of Miami School of Medicine, Miami, FL 33101, USA. 0006-8993/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0 0 0 6 - 8 9 9 3 ( 9 5 ) 0 1 1 3 6 - 6
brain tissue is deprived of oxygen, it rapidly consumes stores of glycogen and ATP, loses its ability to maintain ionic and chemical gradients, then undergoes anoxic depolarization [13,50]. Tolerance of hypoxia in the turtle has a second dimension, as well. In a d d i t i o n to having mechanisms that prevent anoxic depolarization during a hypoxic period, pond turtles have also developed a unique defense to prevent oxidative damage during subsequent reoxygenation of brain tissue: turtle CNS has levels of the antioxidant ascorbic acid (ascorbate) that are 2 - 3 - f o l d higher than in anoxia-intolerant species [41], including mammals [25,31,34,41]. Ascorbate and the thiol tripeptide glutathione (GSH) are the most abundant low molecular mass antioxidants in the CNS [28]. Unlike ascorbate, however, cerebral G S H content is relatively species-invariant [25,41]. Interestingly, ascorbate, in addition to acting directly as an antioxidant [1,3,30,33], has been shown to act as a neuro-
46
M.A. POrez-Pinzdn, M.E. R i c e / B r a i n Research 705 (1995) 45-52
modulator in several neurotransmitter systems [39], which may also contribute to its role as a neuroprotective agent. The difference in ascorbate content between turtle and mammalian CNS [41] is particularly striking in light of the low rate of oxidative metabolism in poikilothermic reptiles [11,15,18,23,54,55,59]. Interspecies comparisons indicate that high ascorbate levels are correlated the anoxia-tolerance of each species, rather than other parameters like metabolic rate [41]. A further difference between turtle and mammalian brain tissue is that turtle brain maintains normal levels of ascorbate [40] and GSH (Druzgal and Rice, unpublished data) for prolonged periods when incubated in vitro in the absence of external sources, whereas mammalian brain tissue rapidly loses cellular ascorbate and GSH in vitro [29,43,46]. Regulation of antioxidant enzymes may also differ between reptile and mammalian brain. Hermes-Lima and Storey [16] recently reported that the activities of glutathione peroxidase and catalase increased in non-neuronal tissue from a freeze-tolerant reptile exposed to sub-freezing conditions. These investigators linked this response to the need for antioxidant protection during reoxygenation after the hypoxia that accompanies freezing [16], which suggests that antioxidant regulation might be an important factor in winter adaptation. In a previous study, we found that basal levels of ascorbate in the isolated turtle cerebellum varied among control groups [40]. Although we did not have the collection history of the animals used in that in vitro study, a possible explanation was that CNS ascorbate levels varied with season, as had been previously reported for the ascorbate content of muscle and other tissues in the frog [51]. The question was important to resolve, because of the possible link between antioxidant regulation and seasonal adaptation in the turtle: turtles undergo prolonged hypoxia during winter hibernation, as well as during periodic dives [14,15,20,21,55]. Indeed, Penney [35] has proposed that the diving capability of turtles is simply a 'spin-off' from adaptive strategies for overwinter hibernation. Metabolic depression accompanies both diving and adaptation to cold temperatures [11,15,18,23,35,55]. Consistent with this, turtles can tolerate anoxic-submergence for several hours at room temperature, whereas this tolerance extends to months at 3°C [14,15]. Emergence from hibernation would be expected to initiate a period of increased metabolic rate in the brain, similar to that which follows a hypoxic dive [50], so that adequate antioxidant protection would be essential throughout the winter, as well as in summer. The goal of the present study was to establish the pattern and temperature-dependence of seasonal variations in ascorbate in the CNS of the pond turtle, Trachemys scripta, with emphasis on the question of whether winter levels of ascorbate remain higher than in mammalian brain. The seasonal dependence of GSH was also assessed. A preliminary report of this research has been presented [37].
2. Materials and methods
2.1. Animals Turtles, Trachemys scripta elegans, were obtained from Kons Scientific (Kenosha, Wisconsin) during summer (July-August) and winter (January-February). The CNS of turtles from five different groups were analyzed for ascorbate and/or GSH content. The groups were: Group name Summer Winter Winter-South Warm Cold
Temperature 23°C 2°C 18°C 23°C l°C
Turtles were collected in the state of Louisiana, then transferred to Wisconsin and kept in an outdoor pond at ambient temperature (supplier information). For most experimental groups, body temperature at the time of dissection was not determined, rather the immediately preceding average temperature for each group was considered to be the best factor for correlation with antioxidant levels. Core temperature (using a rectal probe thermometer) was determined for Cold animals. Summer, Winter, and WinterSouth groups had free access to air for at least the 24 h period during transport. Warm and Cold groups had access to air during the entire re-acclimation period. Summer turtles (9 animals) were obtained in July and August and maintained in tanks with circulating water and a basking platform indoors at room temperature (23°C). Turtles were fed dried krill. Cerebral ascorbate and GSH levels in animals housed indoors for up to several months were not different from those in Summer turtles analyzed immediately after arrival. Winter turtles (7 animals) were shipped from Wisconsin in January and February and CNS ascorbate levels were determined immediately. The average temperature at the supplier's location in Wisconsin was 2°C for that period (National Weather Service). Some turtles (5 animals) were collected in Louisiana and shipped directly to us (Winter-South). These animals were exposed to a milder climate during the same period, with an average temperature of 18°C (National Weather Service). Turtles re-acclimated to warm temperatures (Warm) were shipped from Wisconsin, then maintained at room temperature (23°C) for 3 weeks (4 animals). Summer turtles were acclimated to cold temperatures (Cold) were kept in icy water for 1 week (7 animals).
2.2. Ascorbate and GSH analyses To prepare tissue samples, turtles were quickly decapitated and the whole brain removed. Because we have found gender differences in ascorbate and GSH levels in
M.A. POrez-Pinz6n, M.E. Rice/Brain Research 705 (1995) 45-52
turtle CNS, with 10-15% higher levels in males compared to females [7], the number of animals of each sex was balanced, when possible. The CNS was generally dissected into seven regions: olfactory bulb, cortex, dorsal ventricular ridge (DVR), optic tectum, cerebellum, brain stem and spinal cord, with optic nerve included in some experimental groups. Hemisections of each region were weighed in 1.5 ml microcentrifuge tubes to obtain tissue wet weight, frozen in dry ice, then stored at - 8 0 ° C until analysis was carried out. Total tissue ascorbate content was determined using a reversed phase liquid chromatography (LC) system (BAS 400, Bioanalytical Systems) with a 3 /zm particle Ca8 column (10 cm, Phase II ODS-3, BAS) and 7 /~m guard column (1.5 cm, ODS, BAS). The electrochemical detector was a glassy carbon electrode set at + 0.7 V vs. A g / A g C I . The eluent was 50 mM dibasic sodium phosphate (pH 4.8) in 30% methanol/water and 100 mg 1 1 EDTA, with 150 mg 1-1 myristyl dimethylbenzylammonium chloride (Myris-100, Jame Fine Chemicals, Bound Brook, NJ) as the ion pairing agent. The isocratic pump (Spectra Physics) flow rate was 1 ml rain -1. The eluent was bubbled with argon and then vacuum-filtered. Argon was flushed over the eluent throughout the analysis. Standards were made in ice-cold deoxygenated eluent from 4 mM stock solutions in cold eluent. Samples were sonicated in eluent (1.0 ml) then centrifuged at 12,700 X g for 3 min. The supernatant was decanted and injected (10 /.tl) directly into the HPLC system. To confirm that the single chromatographic peak obtained using a glassy carbon electrode was indeed ascorbate, the specific enzyme, ascorbate acid oxidase (see [42]) was added to selected brain sample homogenates from Summer and Winter turtles. Complete elimination of the peak confirmed ascorbate identity (not illustrated). In experiments when GSH was also determined, the eluent was 20 mM monochloroacetic acid, with 300 mg 1 1 Myris-100 and 100 mg 1-1 EDTA, titrated with NaOH to pH 5. Separation was on a 10 cm Cls reversed phase column (HRA catecholamine column, ESA, Wiggens, MA), with detection using a gold amalgam electrode set at + 0.15 V vs. Ag/AgC1. Ascorbate and GSH eluted in less than 5 min at 1 ml min 1 flow rate [41].
and dry weight for each CNS region in this report. Samples were dissected and frozen on dry ice. The tissue was weighed on a small piece of foil, then dried overnight in a oven at 90-95°C. 2.4. Data analysis
Differences between total tissue contents of ascorbate or GSH for a region in two experimental conditions were analyzed for significance using Student's t-test (e.g. Summer vs. Winter). Differences among CNS regions were analyzed using one-way analysis of variance (ANOVA), followed by a Scheffe's test. Ascorbate and GSH data are expressed as mean + S.E.M., with units of /~mol g - i tissue wet weight. The temperature-dependence of ascorbate levels was estimated using the average environmental temperatures for the turtles. Water content is expressed as percent by weight of tissue wet weight.
3. R e s u l t s
The CNS of pond turtles had significant regional variation in levels of ascorbate and GSH in Summer and in Winter (Figs. 1 and 2), with more pronounced variation in ascorbate than GSH. The general pattern of decreasing anterior-to-posterior levels in ascorbate and GSH content reported previously [41] was seasonally invariant. In both Summer and Winter, the highest values were in olfactory bulb, cortex and DVR, with intermediate levels in optic tectum and cerebellum, and the lowest values in brainstem, spinal cord, and optic nerve. Absolute levels of ascorbate and GSH, on the other hand, were seasonally-dependent throughout the CNS (Figs. 1 and 2), with lower levels in Winter than in Summer. In
Summer
Winter
4
E
8 To determine whether changes in CNS water content contributed to temperature-dependent changes in CNS ascorbate and GSH levels, we determined tissue water content in warm- and cold-acclimated animals. All animals (n = 7) in this study were initially warm-acclimated for at least one month in a thermostatically-controlled room at 20°C, then 3 animals were cold-acclimated in icy water for 2 weeks, while the remaining warm animals were maintained at 20°C. Rectal temperature was taken for each animal immediately before decapitation. Tissue water content was determined as the difference between wet weight
~
=-. 5 E
2.3. Tissue water content
47
3 2
2 1 0 OB
CTX
DVR
OT
CB
BS
SC
ON
Fig. 1. Distribution of ascorbate content in the CNS of ascorbate in Summer and Winter turtles. Significant differences were found between Summer and Winter turtles throughout the CNS (* P < 0.05, * * * P < 0.001). Significant regional variation was preserved in both seasons, as well. CNS regions examined were: olfactory bulb (OB), cortex (CTX), dorsal ventricular ridge (DVR), optic tectum (OT), cerebellum (CB), brainstem (BS), spinal cord (SC), and optic nerve (ON).
M.A. P~rez-Pinz6n, M.E. Rice/Brain Research 705 (1995) 45-52
48 Summer
~
Winter
~
Winter-South
Winter
5 o
E
4
E
g
-£ x:
g
3
1
cO
t i
0
0 OB
CTX
DVR
OT
CB
BS
SC
ON
Fig. 2. Distribution of GSH content in the CNS of Summer and Winter turtles. Significant seasonal differences were found for most CNS regions (* P < 0.05, * * P < 0.01; * * * P < 0.001). Differences in GSH among regions of the CNS and between seasons were generally less pronounced than those for ascorbate (compare with Fig. 1). Abbreviations: see Fig. 1.
general, seasonal differences were greater for ascorbate than for GSH. The average drop in ascorbate levels between Summer and Winter was 33 _+ 1%, taken over all regions examined (n = 8), which was significantly greater than the average 25 _+ 3% fall in GSH levels ( P < 0.01, paired regional comparison). Winter ascorbate levels fell by 30% in olfactory bulb; 36% in cortex; 35% in DVR; 31% in optic tectum; and 27% in the cerebellum. White matter-rich regions, brainstem, spinal cord and optic nerve, also showed significant falls in ascorbate content between the seasons. Brainstem ascorbate levels fell by 38%; spinal cord levels by 37%; and optic nerve by 31%. By contrast, although GSH levels in the CNS were also lower in Winter than in Summer, the differences were less than that for ascorbate, as noted above. Significant changes were seen in several areas, with a fall of 22% in olfactory bulb; 26% in cortex; 20% in DVR; 28% in brainstem; and 42% in spinal cord. The difference between Summer and Winter GSH content did not reach significance in cerebellum, optic tectum, or optic nerve. Because of the potential importance of high ascorbate levels in turtle brain, we followed up these seasonal comparisons with specific studies to explore the influence of temperature on CNS ascorbate content. We first tested whether seasonal differences in ascorbate levels were linked to environmental temperature by comparing the level in turtles acclimatized to winter in different climatic zones of North America (Wisconsin and Louisiana). Winter ascorbate levels in turtles from Louisiana (Winter-South, 19°C) were higher throughout the CNS compared to those in Wisconsin turtles (Winter, 2°C) (Fig. 3), with significant differences in DVR, optic tectum, and cerebellum ( P
