Search tips
Search criteria 


Logo of jcbfmJournal of Cerebral Blood Flow & Metabolism
J Cereb Blood Flow Metab. 2010 April; 30(4): 734–738.
Published online 2010 January 27. doi:  10.1038/jcbfm.2010.9
PMCID: PMC2848708

Differential changes of glutathione levels in astrocytes and neurons in ischemic brains by two-photon imaging


Using two-photon imaging techniques with monochlorobimane as a glutathione (GSH) probe, we investigated GSH levels in both core and penumbra regions of an ischemic brain after middle cerebral artery occlusion. We found that the GSH level significantly decreased in the ischemic core, but increased significantly in the penumbra. Furthermore, we observed a differential change of the GSH levels in neurons and astrocytes in the penumbra. The GSH level in neurons increased significantly whereas it decreased slightly in astrocytes in the penumbra. These findings reveal critical region and cell type-dependent changes of the GSH level in an ischemic brain.

Keywords: astrocyte, GSH, MCAO, neuron, penumbra, stroke


Redox changes and oxidative stress have been implicated in ischemia-caused cell death in the brain. Glutathione has a critical function in maintaining cellular redox status and protecting against oxidative injuries. Determining changes of redox status and GSH levels that occur during ischemic insults is very important in elucidating the pathology of stroke and designing efficient treatment strategies. The majority of the results obtained from early biochemical analyses (Rehncrona and Siesjo, 1979), the most recent electron paramagnetic resonance imaging (Hyodo et al, 2008; Yamato et al, 2009), and magnetic resonance spectroscopy (Lei et al, 2009) have all revealed a reduced global GSH level and a more oxidizing environment in an ischemic brain. However, there are important questions left unanswered. Focal cerebral ischemia results in distinct pathological sites in the brain, which can be classified by an ischemic core and a penumbra. Penumbral cells are potentially salvageable and are potential targets of rescue treatments. At the present time, GSH levels in the ischemic core have not been distinguished from those in the penumbra. Furthermore, changes of GSH levels in different types of cells, particularly neurons and astrocytes, after ischemic insults have not been investigated. Recently, an approach has been developed to map GSH levels in tissues at a cellular level with two-photon microscopy (Sun et al, 2006). We used this technique to image the overall changes of the GSH levels in the penumbra and core of post-ischemic brain slices, and to distinguish changes in GSH levels in neurons and astrocytes.

Materials and methods

Male SD rats weighing between 280 and 310 g were obtained from Charles River (Wilmington, MA, USA). All procedures were approved by the IACUC at UNM HSC and were in accordance with NIH GCULA. For all surgical procedures, the rats were anesthetized by inhalation of 3% isoflurane for induction (1.2% for maintenance) in 70% N2O and 30% oxygen. Rectal temperature was monitored and maintained at 37.5°C. Middle cerebral ischemia/reperfusion was conducted according to Longa et al (1989) with modification as in our previous publication (Liu et al, 2004). Successful middle cerebral artery occlusion was confirmed postmortem by 2,3,5-triphenyltetrazolium chloride staining and through observation of behavioral changes.

After ischemia/reperfusion, brains were rapidly removed from the skull and placed in ice-cold cutting solution. Coronal sections (350 μm) were cut using a Vibratome (Technical Products, St Louis, MO, USA). The slices were placed into artificial cerebrospinal fluid equilibrated with 95% O2 and 5% CO2 for 1 h of recovery at room temperature. For identifying astrocytes, we incubated slices in 1 μmol/L SR101 in oxygenated artificial cerebrospinal fluid for 30 mins at 35°C for optimal staining. After SR101 staining, the bathing solution was changed to oxygenated artificial cerebrospinal fluid, and slices were held at room temperature until optical measurements were begun. For imaging, we transferred individual slices to a recording chamber and perfused with oxygenated artificial cerebrospinal fluid at 2 mL/min at room temperature. Monochlorobimane (Sigma, St Louis, MO, USA) was added to the perfusion system at a final concentration of 60 μmol/L. Costaining of propidium iodide (PI, 10 μmol/L) with MCB was performed to identify dead or dying cells. All the imaging process was performed at room temperature to minimize MCB conjugate efflux (Sun et al, 2006).

GSH–MCB fluorescence was visualized using an Olympus BX 51WI upright microscope water-immersion LUMPlan FL/IR × 60/0.90 or × 20/0.95 W objective (Tokyo, Japan). Excitation was provided by a Bio-Rad Radiance 2100 MP laser scan unit powered by a Spectra Physics Tsunami/Millennia VI laser combination (Hercules, CA, USA). Band-pass-filtered epifluorescence was collected by the photomultiplier tubes of the Radiance system. Images were acquired using Zeiss Lasersharp 2000 software routines. Fluorescence was collected using PMT1 for the GSH–MCB signal (488 to 565 nm), and PMT2 for the SR101 and PI signal (620 to 645 nm). All images were taken at 50 to 60 μm below the surface to avoid slicing effect and minimize perfusion (see Supplementary information for detailed rational). Background subtraction was performed for each experiment. Background signal in neurons was 8.3±1.2 a.u. and in astrocytes was 8.5±1.4 a.u.

2,3,5-Triphenyltetrazolium chloride staining was performed on adjacent slices to localize normal tissue, ischemic core, and penumbra (see Supplementary information for detail). Determination of GSH fluorescence within individual cells was made in offline analyses using NIH ImageJ (Bethesda, MD, USA). Rectangular areas of interest (AOIs) were positioned within individual cell bodies that showed GSH, but not PI fluorescence. Background corrected data from 10 to 20 cell bodies per AOI were averaged to give GSH levels in live cells. GSH–MCB fluorescence values were converted to intracellular GSH concentration (mmol/L) using a calibration curve (Figure 1A) according to a published method (Meyer and Fricker, 2000; Sun et al, 2006; see Supplementary information for detailed). For each animal we averaged the GSH levels from three slices and then calculated the average from 10 animals. For statistical analyses, we used Student's t-test with SigmaPlot 10.0 (San Jose, CA, USA). Before t-testing, we tested and confirmed the normality of the samples. Data were presented in box-whisker plots.

Figure 1
Alterations in GSH levels in the penumbra and the ischemic core measured by two-photon microscopy. (A) Standard curve of GSH concentration (see Supplementary information for detailed method). (B) 2,3,5-Triphenyltetrazolium chloride-stained rat brain coronal ...


Figure 1B is a typical 2,3,5-triphenyltetrazolium chloride-stained brain slice and shows selected locations for imaging. To measure cellular GSH levels in ischemic brains, it is critical that GSH is completely reacted with MCB. Kinetic monitoring confirmed complete reaction in the core, penumbra, and normal tissues (Figure 1C, see Supplementary information for detail). Figure 1D shows representative GSH–MCB fluorescence images in a brain slice from an animal after 90 min unilateral occlusion and 24 h reperfusion. The locations of the sampled regions were selected to represent ischemic core (AOI #1*), penumbra (AOI #2*), and normal tissue (AOI #3*). Symmetrical locations at the contralateral side were selected as normal controls. GSH mapping at the control sites (Figure 1D, AOIs #1, 2, 3, and 3*) showed that MCB strongly labeled a major subpopulation of cell bodies in cortex with less fluorescence seen in neuropil. The GSH level in the four control sites was 0.47±0.07 mmol/L, and there was no significant difference between them (Figure 1E). The GSH level in the core significantly decreased to 0.25±0.03 mmol/L (53%±5.3% of the control level, Figure 1D AOI #1* and 1E). It is noteworthy to point out that average GSH level in the core should be lower than 0.25 mmol/L as the value was average from PI-negative cells in the core and most of the cells in the core were PI-positive (see Supplementary information). Surprisingly, the GSH level (0.68±0.08 mmol/L) in the penumbra was significantly higher than the control with a 147%±7.4% increase.

The fluorescence in the non-cell body space of penumbra was brighter than other images in Figure 1D. This difference prompted us to look at the images at a higher magnification. Figure 1F shows GSH–MCB fluorescence images in the contralateral side and the penumbra taken with a × 60 objective. These images clearly show that the increase of fluorescence in non-cell body space in penumbra over the contralateral side. This indicates that the GSH level increased not only in cell bodies, but also in cellular processes in the penumbra.

The above results showed that the GSH levels were elevated in the penumbra after 90 min ischemia and 24 h reperfusion. Because GSH fluorescence was particularly bright in some cell bodies, we set out to identify the cell types responsible. We performed double staining of live tissue with MCB and SR101, the latter a marker for astrocytes in live tissue (Nimmerjahn et al, 2004). Neuron somata can be discriminated from astrocytes by a larger, pyramidal- or oval-shaped morphology and the absence of SR101 labeling. As there was no significant difference in the GSH levels in the four selected control, only was one control site (AOI #2 in Figure 1) used to be compared with penumbra and core. Figure 2A (AOIs #1 to 3) shows that in the contralateral cortex, SR101-positive cells (astrocytes, arrows) show significantly higher GSH concentration than neurons. Figures 2B and 2C show that neurons (0.39±0.04 mmol/L) had significantly less GSH than astrocytes (0.8±0.08) in control tissues. This observation is highly in agreement with a previous report where a higher GSH level was observed in astrocytes than in neurons in normal brain slices (Sun et al, 2006). A greater difference between neuron and astrocyte GSH levels in Sun's report may result from different animal age and different imaging depth (see discussion in Supplementary information).

Figure 2
Changes in the GSH levels in astrocytes and neurons in penumbra after 90 min ischemia and 24 h reperfusion. (A) Representative two-photon fluorescence images of GSH–MCB labeled cortex in a coronal brain slice from an ischemic rat ...

In contrast to normal tissue, neurons in the penumbra showed a robust increase (62%) in the GSH level (from 0.39±0.03 to 0.63±0.07 mmol/L, Figure 2C), whereas the GSH level in astrocytes decreased slightly (from 0.8±0.08 to 0.73±0.07 mmol/L, Figure 2B). In the ischemic core, GSH content of astrocytes dropped to 36% of basal level (0.29±0.03 mmol/L, Figure 2B) whereas in neurons it decreased to 52% of the control level (0.2±0.01 mmol/L, Figure 2C). These results revealed that there were significant differences in cellular responses to ischemic insult between neurons and astrocytes, in term of the GSH level. The elevation of the GSH level in neurons in the penumbra possibly has a function in preventing greater loss of neurons.


A reduced level of GSH has typically been reported in ischemic brains (Hyodo et al, 2008; Lei et al, 2009; Rehncrona and Siesjo, 1979; Shivakumar et al, 1995; Yamato et al, 2009) using techniques that do not give spatially resolved measurements. In our study, we were able to differentiate GSH levels in the penumbra and the ischemic core. The GSH levels in the ischemic core decrease dramatically. The GSH levels in the penumbra, however, increase significantly, associated with increased neuronal GSH levels.

The difference in the GSH levels between the penumbra and the core may reflect the ambient conditions in the two regions during the ischemic insult and early phases of reperfusion. One of these conditions is that the decrease in nutrient supplies such as glucose and oxygen is mild in the penumbra and severe in the core. Glucose has a critical function in maintaining cellular GSH level under normoxic conditions by providing nicotinamide adenine dinucleotide phosphate through the pentose phosphate pathways, as well as in hypoxic conditions (Gupte et al, 2006; White et al, 1988). This may explain, at least in part, the observed increased GSH levels in the penumbra that is hypoxic. Strong loss of GSH in the ischemic core presumably reflects irreversible cellular damage. The difference can also be partly explained by the balance of synthesis and consumption. In the core, synthesis of GSH is largely eliminated. In the penumbral neurons, the viable GSH synthesis pathways, available substrates, and stimulation by the increased oxidative stress would suggest a higher synthesis. In addition, hypoxia-inducible factor 1, induced in the penumbra, may have a direct or indirect function in increasing GSH levels in the penumbra (Guo et al, 2008, 2009).

There is heterogeneity in the GSH levels in neurons and glial cells in normal brain tissue, with astrocytes in general exhibiting much higher levels of GSH, as shown in this report and the previous report (Sun et al, 2006). Our results show that under ischemic condition there are differential changes of the GSH level in neurons and astrocytes. The mechanism for the increased GSH levels in neurons in the penumbra is not clear. However, the following factors may be involved. First, neuronal GSH maintenance depends on astrocytes (Dringen et al, 1999). Under hypoxic conditions, GSH support to neurons from astrocytes may be increased. Second, hypoxia-inducible factor 1 expression may be different in hypoxic neurons and astrocytes, which would result in different regulations of GSH levels in the two cell types. Third, glutamate levels in the two cell types may be different and thus affect the GSH levels as glutamate affects GSH synthesis and oxidation in ischemia/reperfusion (Swanson et al, 2004). Nevertheless, future studies are needed to clarify the molecular mechanism of GSH increase in neurons in the penumbra.

In conclusion, this report highlights a significant and critical distinction in the GSH levels between ischemic core and penumbra. The GSH level increased in the penumbra, compared to contralateral side, whereas it decreased in the ischemic core. Furthermore, it revealed important and differentiated changes of GSH levels in astrocytes and neurons in both the ischemic core and the penumbra.


Supplementary Information accompanies the paper on the Journal of Cerebral Blood Flow & Metabolism website (

Supplementary Material

Supplementary Information


  • Dringen R, Pfeiffer B, Hamprecht B. Synthesis of the antioxidant glutathione in neurons: supply by astrocytes of CysGly as precursor for neuronal glutathione. J Neurosci. 1999;19:562–569. [PubMed]
  • Guo S, Bragina O, Xu Y, Cao Z, Chen H, Zhou B, Morgan M, Lin Y, Jiang BH, Liu KJ, Shi H. Glucose up-regulates HIF-1 alpha expression in primary cortical neurons in response to hypoxia through maintaining cellular redox status. J Neurochem. 2008;105:1849–1860. [PubMed]
  • Guo S, Miyake M, Liu KJ, Shi H. Specific inhibition of hypoxia inducible factor 1 exaggerates cell injury induced by in vitro ischemia through deteriorating cellular redox environment. J Neurochem. 2009;108:1309–1321. [PMC free article] [PubMed]
  • Gupte SA, Okada T, McMurtry IF, Oka M. Role of pentose phosphate pathway-derived NADPH in hypoxic pulmonary vasoconstriction. Pulm Pharmacol Ther. 2006;19:303–309. [PubMed]
  • Hyodo F, Chuang KH, Goloshevsky AG, Sulima A, Griffiths GL, Mitchell JB, Koretsky AP, Krishna MC. Brain redox imaging using blood-brain barrier-permeable nitroxide MRI contrast agent. J Cereb Blood Flow Metab. 2008;28:1165–1174. [PMC free article] [PubMed]
  • Lei H, Berthet C, Hirt L, Gruetter R. Evolution of the neurochemical profile after transient focal cerebral ischemia in the mouse brain. J Cereb Blood Flow Metab. 2009;29:811–819. [PubMed]
  • Liu S, Shi H, Liu W, Furuichi T, Timmins GS, Liu KJ. Interstitial pO2 in ischemic penumbra and core are differentially affected following transient focal cerebral ischemia in rats. J Cerebr Blood Flow Metab. 2004;24:343–349. [PubMed]
  • Longa EZ, Weinstein PR, Carlson S, Cummins R. Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke. 1989;20:84–91. [PubMed]
  • Meyer AJ, Fricker MD. Direct measurement of glutathione in epidermal cells of intact Arabidopsis roots by two-photon laser scanning microscopy. J Microsc. 2000;198:174–181. [PubMed]
  • Nimmerjahn A, Kirchhoff F, Kerr JN, Helmchen F. Sulforhodamine 101 as a specific marker of astroglia in the neocortex in vivo. Nat Methods. 2004;1:31–37. [PubMed]
  • Rehncrona S, Siesjo BK. Cortical and cerebrospinal fluid concentrations of reduced and oxidized glutathione during and after cerebral ischemia. Adv Neurol. 1979;26:285–286. [PubMed]
  • Shivakumar BR, Kolluri SV, Ravindranath V. Glutathione and protein thiol homeostasis in brain during reperfusion after cerebral ischemia. J Pharmacol Exp Ther. 1995;274:1167–1173. [PubMed]
  • Sun X, Shih AY, Johannssen HC, Erb H, Li P, Murphy TH. Two-photon imaging of glutathione levels in intact brain indicates enhanced redox buffering in developing neurons and cells at the cerebrospinal fluid and blood-brain interface. J Biol Chem. 2006;281:17420–17431. [PubMed]
  • Swanson RA, Ying W, Kauppinen TM. Astrocyte influences on ischemic neuronal death. Curr Mol Med. 2004;4:193–205. [PubMed]
  • White CW, Jackson JH, McMurtry IF, Repine JE. Hypoxia increases glutathione redox cycle and protects rat lungs against oxidants. J Appl Physiol. 1988;65:2607–2616. [PubMed]
  • Yamato M, Shiba T, Yamada KI, Watanabe T, Utsumi H. Noninvasive assessment of the brain redox status after transient middle cerebral artery occlusion using overhauser-enhanced magnetic resonance imaging. J Cereb Blood Flow Metab. 2009;29:1655–1664. [PubMed]

Articles from Journal of Cerebral Blood Flow & Metabolism are provided here courtesy of SAGE Publications