- Email: [email protected]
Microdialysis perfusion increases the sensitivity of rat striatal neurons to ischemic insult
Brain Research, 578 (1992) 339-341 © 1992 Elsevier Science Publishers B.V. All rights reserved. 0006-8993/92/$05.00
339
BRES 25138
Microdialysis perfusion increases the sensitivity of rat striatal neurons to ischemic insult Lee A. Phebus, Ron E. Mincy and James A. Clemens Central Nervous System Research, The Lilly Research Laboratories, Eli Lilly and Co., Indianapolis, IN 46285 (USA)
(Accepted 21 January 1992) Key words: Calcium; Cerebral ischemia; Four-vessel occlusion; Microdialysis; Rat; Striatum
Brief periods of global forebrain ischemia which never produced striatal cell loss in control, non-dialyzed striatum resulted in significant cell loss in the contralateral, dialyzed striatum. No striatal cell loss was seen following these short periods of ischemia in animals implanted with dialysis probes that were not perfused. Dialysis perfusion alters striatal extracellular fluid composition in such a way as to render this tissue more sensitive to brief ischemic insult. In order to determine whether the release of a particular neurochemical during ischemia was tightly associated with striatal cell death, we examined the release of these compounds during periods of ischemia known to be too short to produce striatal ischemic cell lOSS3'6. To our surprise, striatal cell death was seen, localized solely to areas influenced by dialysis perfusion, following periods of ischemia too short to damage the contralateral non-dialyzed striatum. The interaction between dialysis perfusion and striatal ischemic damage was examined further in this study. Wistar rats, anesthetized with Metofane, were prepared for the 4-vessel occlusion (4-VO) model of global forebrain ischemia using the method of Pulsinelli and Brierley 2'6. At the same time, loop-style dialysis probes were permanently placed, unilaterally, in their dorsolateral striata 1'5. The dialyzing loop of the probe was 3 mm long with a width of 0.7 mm in the plane of the loop and 0.3 mm on its narrowest axis. One day later, after recovery from anesthesia, these rats were placed in a cylindrical chamber and connected to a syringe pump which perfused the dialysis probes with artificial cerebro-spinal fluid (CSF) at a rate of 1 / d per min. The artificial CSF had a composition of 150 mM NaCI, 3 mM KC1, 1.7 mM CaC12 and 0.9 mM MgC12 with a pH of 7.4. In some cases 0 and in other cases 10 mM CaCI2 was used with a compensatory change in NaCI concentration. Probes were perfused from 1 h before ischemia until 2 h after reperfusion. Either 10 or 15 min of global cerebral ischemia was induced. Groups used in the 10 min ischemia experiment included sham ischemia, and ischemia with non-perfused or artificial CSF-perfused dialysis probes.
The 15 rain ischemia experiment also included groups whose dialysis probes were perfused with artificial CSF containing 0 or 10 mM calcium. Global forebrain ischemia was induced as previously described 2. Either 10 or 15 min after the onset of ischemia, the brain was allowed to reperfuse, terminating the ischemia. Only rats that remained non-responsive for the entire ischemic period were used in this study. During cerebral ischemia and the first 30 min of reperfusion, the animal's core temperature was maintained at 37°C. 2 h after reperfusion, rats were disconnected from the syringe pump and placed in individualp cages with free access to food and water. 24 h after ischemia, the rats were killed and their brains were removed, frozen and sectioned on a cryostat. Every fifth 32/~m section was mounted on a microscope slide and stained with Cresyl violet. When the dialysis probe tract was visible, more frequent sections were taken. Using image analysis, the area of striatal cell loss, in square millimeters, was calculated for each section. The section with the largest area of striatal damage was used to represent the damage in that particular animal. Data were analyzed using a one-way analysis of variance with a post-hoc comparison of confidence intervals (Minitab, State College, PA). A probability level less than 0.05 was considered statistically significant. In the sham ischemia group, the striatal damage resulting from insertion of the dialysis probe and its perfusion with artificial CSF was 0.58 + 0.09 mm 2 (mean + S.E.M., n = 8). This is essentially the area of the probe tract through striatal tissue.
Correspondence: L.A. Phebus, Dept. Mc-907, Lilly Corporate Center, Indianapolis, IN 46285-0815, USA.
340 4
a Cr
0" or)
m
q¢ m
2
E
E
m r~
t~
~6 .e E
SHAM A-CSF
4-VO N.P.
SHAM A-CSF
4-VO A-CSF
4-VO N.P.
4-VO A-CSF
Fig. 1. Striatal damage, measured as described in the text, in animals previously implanted with dialysis probes and undergoing sham ischemia (SHAM) or 10 rain of 4-VO ischemia (4-VO). The dialysis probes were either perfused with artifieal CSF (A-CSF) or not perfused (N.E). *P < 0.05 vs. SHAM and N.E groups.
Fig. 2. Striatal damage in animals previously implanted with dialysis probes and undergoing sham ischemia (SHAM) or 15 min of 4-VO ischemia (4-VO) with non-perfused probes (N.E) or artificial CSF perfused probes (A-CSF). *P < 0.05 vs. SHAM and N.E groups.
10 min of 4 - v o ischemia with artificial CSF perfusion resulted in 3.19 + 0.64 m m 2 of striatal damage (n = 16). The damaged area was usually distributed symmetrically around the dialyzing surface of the probe. 10 min of ischemia with a non-perfused probe produced 0.74 + 0.12 mm 2 of striatal damage (n = 9), not significantly different from the sham group. The combination of 10 min of ischemia and artificial CSF perfusion of the dialysis probe resulted in significantly greater damage than the sham ischemia with artificial CSF perfusion or the ischemic but not perfused groups (Fig. 1). In none of these groups was there any detectable striatal damage in the non-implanted, contralateral hemisphere. 15 min of 4-VO ischemia in animals whose dialysis probes were perfused with artifical CSF resulted in 2.45 + 0.40 mm 2 of striatal damage (n = 16). Identical ischemia in animals with non-perfused dialysis probes produced significantly less striatal damage (0.68 + 0.08, n = 10) (Fig. 2). These results are essentially identical to those found in the 10 min ischemia experiment. Perfusion with artificial CSF containing 0 mM calcium increased the amount of striatal damage but not significantly (4.76 + 0.97, n = 10). Perfusion with 10 mM calcium significantly increased the striatal damage produced by 15 min of 4-VO ischemia with perfusion (6.46 _+ 0.82, n = 23) (Fig. 3). There was never any cell loss detected in the contralateral, non-implanted striatum as a result of the 15 min of 4-VO ischemia. Following periods of 10 or 15 min of 4-VO ischemia, there was never any striatal cell loss in the control, nonimplanted striatum. This fits well with the established fact that it takes about 20 min of 4-VO ischemia to produce striatal damage in rats 3'6. In the implanted striatum, ischemic cell loss was seen only when the dialysis
probe was perfused. This argues that neither the mechanical damage produced by the insertion of the probe nor the response of striatal tissue to the physical presence of the probe's various components was responsible for the exacerbation of striatal ischemic damage. The observation that the cell loss was distributed about the dialyzing surface of the probe argues that alterations in the local extracellular environment produced by microdialysis perfusion resulted in heightened striatal sensitivity to ischemia. The composition of the artificial CSF used in this study is typical of that used in the microdialysis literature 4. The osmolarity and pH are appropriate for brain extracellular fluid but the ionic composition is a compromise between simplicity of preparation and
:E o"
a; m
E r~
m
.o
A-CSF
HIGH CA
NO CA
Fig. 3. Effect of varying the perfusate calcium concentration on the striatal damage produced by 15 min of 4-VO ischemia with dialysis perfusion. A-CSF, artificial CSF perfused; HIGH CA, 10 mM calcium artificial CSF; NO CA, calcium-free artificial CSF perfusion. *P < 0.05 vs. artificial CSF (A-CSF) group.
341
the artificial CSF used in this study was 158 mM, which is higher than the approximately 136 m M found in natural CSE Perfusion with this solution would increase local extracellular concentrations of chloride ions, perhaps predisposing surrounding tissue to ischemic damage. Alternatively, since CSF is a complex solution containing hundreds of small molecules, it is also possible that the removal of some protective factor by dialysis perfusion is responsible for the heightened sensitivity of dialyzed striatal tissue to ischemic insult. The persistence of stri-
atal damage when the dialysis probe was perfused with artificial CSF containing no calcium argues against calcium overload as the major mechanism of dialysis exacerbation of striatal ischemic damage. These results may provide some insight into the mechanism of ischemia-induced striatal cell death. Perfusion of the dialysis probe may deplete striatal extracellular fluid of factors that protect, or it may add factors that compromise striatal neurons during ischemia. In either case, the pathophysiology of the ischemic insult is altered by the microdialysis sampling technique.
1 Clemens, J.A. and Phebus, L.A., Brain dialysis in conscious rats confirms in vivo electrochemical evidence that dopaminergic stimulation releases ascorbate, Life Sci., 35 (1984) 671-677. 2 Clemens, J.A. and Phebus, L.A., Dopamine depletion protects striatal neurons from ischemia-induced cell death, Life Sci., 42 (1988) 707-713. 3 Ginsberg, M.D. and Busto, R., Rodent models of cerebral ischemia, Stroke, 20 (1989) 1627-1632. 4 Osborne, P.G., O'Connor, W.T. and Ungerstedt, U., Effect of
varying the ionic concentration of microdialysis perfusate on basal striatal dopamine levels in awake animals, J. Neurochem., 56 (1991) 452-456. 5 Phebus, L.A. and Clemens, J.A., Effects of transient global cerebral ischemia on extracellular dopamine, serotonin and their metabolites, Life Sci, 44 (1989) 1335-1342. 6 Pulsinelli, W.A. and Brierley, J.B., A new model of bilateral hemispheric ischemia in the unanesthetized rat, Stroke, 10 (1979) 267-272.
the actual complexity of true CSE The chloride level in
339
BRES 25138
Microdialysis perfusion increases the sensitivity of rat striatal neurons to ischemic insult Lee A. Phebus, Ron E. Mincy and James A. Clemens Central Nervous System Research, The Lilly Research Laboratories, Eli Lilly and Co., Indianapolis, IN 46285 (USA)
(Accepted 21 January 1992) Key words: Calcium; Cerebral ischemia; Four-vessel occlusion; Microdialysis; Rat; Striatum
Brief periods of global forebrain ischemia which never produced striatal cell loss in control, non-dialyzed striatum resulted in significant cell loss in the contralateral, dialyzed striatum. No striatal cell loss was seen following these short periods of ischemia in animals implanted with dialysis probes that were not perfused. Dialysis perfusion alters striatal extracellular fluid composition in such a way as to render this tissue more sensitive to brief ischemic insult. In order to determine whether the release of a particular neurochemical during ischemia was tightly associated with striatal cell death, we examined the release of these compounds during periods of ischemia known to be too short to produce striatal ischemic cell lOSS3'6. To our surprise, striatal cell death was seen, localized solely to areas influenced by dialysis perfusion, following periods of ischemia too short to damage the contralateral non-dialyzed striatum. The interaction between dialysis perfusion and striatal ischemic damage was examined further in this study. Wistar rats, anesthetized with Metofane, were prepared for the 4-vessel occlusion (4-VO) model of global forebrain ischemia using the method of Pulsinelli and Brierley 2'6. At the same time, loop-style dialysis probes were permanently placed, unilaterally, in their dorsolateral striata 1'5. The dialyzing loop of the probe was 3 mm long with a width of 0.7 mm in the plane of the loop and 0.3 mm on its narrowest axis. One day later, after recovery from anesthesia, these rats were placed in a cylindrical chamber and connected to a syringe pump which perfused the dialysis probes with artificial cerebro-spinal fluid (CSF) at a rate of 1 / d per min. The artificial CSF had a composition of 150 mM NaCI, 3 mM KC1, 1.7 mM CaC12 and 0.9 mM MgC12 with a pH of 7.4. In some cases 0 and in other cases 10 mM CaCI2 was used with a compensatory change in NaCI concentration. Probes were perfused from 1 h before ischemia until 2 h after reperfusion. Either 10 or 15 min of global cerebral ischemia was induced. Groups used in the 10 min ischemia experiment included sham ischemia, and ischemia with non-perfused or artificial CSF-perfused dialysis probes.
The 15 rain ischemia experiment also included groups whose dialysis probes were perfused with artificial CSF containing 0 or 10 mM calcium. Global forebrain ischemia was induced as previously described 2. Either 10 or 15 min after the onset of ischemia, the brain was allowed to reperfuse, terminating the ischemia. Only rats that remained non-responsive for the entire ischemic period were used in this study. During cerebral ischemia and the first 30 min of reperfusion, the animal's core temperature was maintained at 37°C. 2 h after reperfusion, rats were disconnected from the syringe pump and placed in individualp cages with free access to food and water. 24 h after ischemia, the rats were killed and their brains were removed, frozen and sectioned on a cryostat. Every fifth 32/~m section was mounted on a microscope slide and stained with Cresyl violet. When the dialysis probe tract was visible, more frequent sections were taken. Using image analysis, the area of striatal cell loss, in square millimeters, was calculated for each section. The section with the largest area of striatal damage was used to represent the damage in that particular animal. Data were analyzed using a one-way analysis of variance with a post-hoc comparison of confidence intervals (Minitab, State College, PA). A probability level less than 0.05 was considered statistically significant. In the sham ischemia group, the striatal damage resulting from insertion of the dialysis probe and its perfusion with artificial CSF was 0.58 + 0.09 mm 2 (mean + S.E.M., n = 8). This is essentially the area of the probe tract through striatal tissue.
Correspondence: L.A. Phebus, Dept. Mc-907, Lilly Corporate Center, Indianapolis, IN 46285-0815, USA.
340 4
a Cr
0" or)
m
q¢ m
2
E
E
m r~
t~
~6 .e E
SHAM A-CSF
4-VO N.P.
SHAM A-CSF
4-VO A-CSF
4-VO N.P.
4-VO A-CSF
Fig. 1. Striatal damage, measured as described in the text, in animals previously implanted with dialysis probes and undergoing sham ischemia (SHAM) or 10 rain of 4-VO ischemia (4-VO). The dialysis probes were either perfused with artifieal CSF (A-CSF) or not perfused (N.E). *P < 0.05 vs. SHAM and N.E groups.
Fig. 2. Striatal damage in animals previously implanted with dialysis probes and undergoing sham ischemia (SHAM) or 15 min of 4-VO ischemia (4-VO) with non-perfused probes (N.E) or artificial CSF perfused probes (A-CSF). *P < 0.05 vs. SHAM and N.E groups.
10 min of 4 - v o ischemia with artificial CSF perfusion resulted in 3.19 + 0.64 m m 2 of striatal damage (n = 16). The damaged area was usually distributed symmetrically around the dialyzing surface of the probe. 10 min of ischemia with a non-perfused probe produced 0.74 + 0.12 mm 2 of striatal damage (n = 9), not significantly different from the sham group. The combination of 10 min of ischemia and artificial CSF perfusion of the dialysis probe resulted in significantly greater damage than the sham ischemia with artificial CSF perfusion or the ischemic but not perfused groups (Fig. 1). In none of these groups was there any detectable striatal damage in the non-implanted, contralateral hemisphere. 15 min of 4-VO ischemia in animals whose dialysis probes were perfused with artifical CSF resulted in 2.45 + 0.40 mm 2 of striatal damage (n = 16). Identical ischemia in animals with non-perfused dialysis probes produced significantly less striatal damage (0.68 + 0.08, n = 10) (Fig. 2). These results are essentially identical to those found in the 10 min ischemia experiment. Perfusion with artificial CSF containing 0 mM calcium increased the amount of striatal damage but not significantly (4.76 + 0.97, n = 10). Perfusion with 10 mM calcium significantly increased the striatal damage produced by 15 min of 4-VO ischemia with perfusion (6.46 _+ 0.82, n = 23) (Fig. 3). There was never any cell loss detected in the contralateral, non-implanted striatum as a result of the 15 min of 4-VO ischemia. Following periods of 10 or 15 min of 4-VO ischemia, there was never any striatal cell loss in the control, nonimplanted striatum. This fits well with the established fact that it takes about 20 min of 4-VO ischemia to produce striatal damage in rats 3'6. In the implanted striatum, ischemic cell loss was seen only when the dialysis
probe was perfused. This argues that neither the mechanical damage produced by the insertion of the probe nor the response of striatal tissue to the physical presence of the probe's various components was responsible for the exacerbation of striatal ischemic damage. The observation that the cell loss was distributed about the dialyzing surface of the probe argues that alterations in the local extracellular environment produced by microdialysis perfusion resulted in heightened striatal sensitivity to ischemia. The composition of the artificial CSF used in this study is typical of that used in the microdialysis literature 4. The osmolarity and pH are appropriate for brain extracellular fluid but the ionic composition is a compromise between simplicity of preparation and
:E o"
a; m
E r~
m
.o
A-CSF
HIGH CA
NO CA
Fig. 3. Effect of varying the perfusate calcium concentration on the striatal damage produced by 15 min of 4-VO ischemia with dialysis perfusion. A-CSF, artificial CSF perfused; HIGH CA, 10 mM calcium artificial CSF; NO CA, calcium-free artificial CSF perfusion. *P < 0.05 vs. artificial CSF (A-CSF) group.
341
the artificial CSF used in this study was 158 mM, which is higher than the approximately 136 m M found in natural CSE Perfusion with this solution would increase local extracellular concentrations of chloride ions, perhaps predisposing surrounding tissue to ischemic damage. Alternatively, since CSF is a complex solution containing hundreds of small molecules, it is also possible that the removal of some protective factor by dialysis perfusion is responsible for the heightened sensitivity of dialyzed striatal tissue to ischemic insult. The persistence of stri-
atal damage when the dialysis probe was perfused with artificial CSF containing no calcium argues against calcium overload as the major mechanism of dialysis exacerbation of striatal ischemic damage. These results may provide some insight into the mechanism of ischemia-induced striatal cell death. Perfusion of the dialysis probe may deplete striatal extracellular fluid of factors that protect, or it may add factors that compromise striatal neurons during ischemia. In either case, the pathophysiology of the ischemic insult is altered by the microdialysis sampling technique.
1 Clemens, J.A. and Phebus, L.A., Brain dialysis in conscious rats confirms in vivo electrochemical evidence that dopaminergic stimulation releases ascorbate, Life Sci., 35 (1984) 671-677. 2 Clemens, J.A. and Phebus, L.A., Dopamine depletion protects striatal neurons from ischemia-induced cell death, Life Sci., 42 (1988) 707-713. 3 Ginsberg, M.D. and Busto, R., Rodent models of cerebral ischemia, Stroke, 20 (1989) 1627-1632. 4 Osborne, P.G., O'Connor, W.T. and Ungerstedt, U., Effect of
varying the ionic concentration of microdialysis perfusate on basal striatal dopamine levels in awake animals, J. Neurochem., 56 (1991) 452-456. 5 Phebus, L.A. and Clemens, J.A., Effects of transient global cerebral ischemia on extracellular dopamine, serotonin and their metabolites, Life Sci, 44 (1989) 1335-1342. 6 Pulsinelli, W.A. and Brierley, J.B., A new model of bilateral hemispheric ischemia in the unanesthetized rat, Stroke, 10 (1979) 267-272.
the actual complexity of true CSE The chloride level in