glutamate turnover and energy metabolism in brain injury

ACTA

UNIVERSITATIS

UPSALIENSIS

UPPSALA

2008

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 329

Glutamate Turnover and EnergyMetabolism in Brain Injury

Clinical and Experimental Studies

CAROLINA SAMUELSSON

ISSN 1651-6206ISBN 978-91-554-7155-2urn:nbn:se:uu:diva-8630

�� !�� "� #$� �� %�� &�� #� #$$' �� $()*+ ,� �-� �� , �� , �-��-� .%�� ,&��/0 1-� �� 2�� 2��-0

��

�� 30 #$$'0 �� 1�� 4�� &�� 5�� 6 ��0 3�� 4�� 0 7�� 0 �� 8#(0 9* ��0 ��06�5: (9'�(*�++;�9*++�#0

�� -� �� 2-��- �� 0 �� ,�� 0 1-�� 0 �� 2�� ,�� 88 �� 2��- �� -��-��-�� .�7�/0 6�� -�� -� ,�� ,�� , �-� �� <1�* 2�� *�+ �� ,��2�� 0 �� 2��-2 �-�� -� �<1�* �� 2�� 2��- �� 0 1-� �� -��0 7�� 8=$$ �� -�� ,�� 2�� ,�� -�� -� �� >�� .<>�/ �� 2��- �-��>�� .� ?�$0==/0 1-�� 2�� -� �� 2��<>� �� .�?$0='/ �� 2�� .�?$0=+/0 �� .�?$0+#/0 4�� ,�� ,�� <>�@;$� 2�� 2��- -��- �� 0 �� ,�� 0 4�� ,�� 2�� 2��- �� -�� ,�� .�6:�/ � �� ,�� ,�� ,�� ,�� -�� <>�@;$ �� -�� -�� -�� ,�� 7�0 �6:�� 2�� <>� �� 0 �� 2��- �-� �� ,�� .�?$0#+� �?$0#;/0 �� -� ��>�� 2��- �-� �� .�?�$0#(� �?$0;$/0 �� 2�� , �-� �� ,��0 6�� ,�� 2��- �� -��-�� 0

� �� <1�*� �� ,�� -�� -��-�� -��

�� ! "#$� �� %&'"()*��

A 3�� #$$'

6��: *=+*�=#$=6�5: (9'�(*�++;�9*++�#��)�)��)��)��'=8$ .-��)>>��0��0��>��B��?��)�)��)��)��'=8$/

Till mina föräldrar

List of Papers

This thesis is based on the following papers, which will be referred to in the text by their Roman numerals:

I Samuelsson C, Kumlien E, Flink R, Lindholm D, Ronne-

Engström E. Decreased cortical levels of astrocytic glutamate transport protein GLT-1 in a rat model of posttraumatic epi-lepsy. Neuroscience Letters 2000 Aug 11;289(3):185-8

II Samuelsson C, Kumlien E, Elfving Å, Lindholm D, Ronne-

Engström E. The effects of PBN (phenyl-butyl-nitrone) on GLT-1 levels and on the extracellular levels of amino acids and energy metabolites in a model of iron-induced posttrau-matic epilepsy. Epilepsy Research 2003 Oct;56(2-3):165-73

III Samuelsson C, Hillered L, Zetterling M, Enblad P, Hesselager G,

Ryttlefors M, Kumlien E, Lewén A, Marklund N, Nilsson P, Salci K, Ronne-Engström E. Cerebral glutamine and gluta-mate levels in relation to compromised energy metabolism – a microdialysis study in subarachnoidal hemorrhage pa-tients. J Cereb Blood Flow Metab. 2007 Jul;27(7):1309-17

IV Samuelsson C, Hillered L, Kumlien E, Enblad P, Ronne-

Engström E. Microdialysis patterns in subarachnoid hemor-rhage patients with focus on ischemic events and brain inter-stitial glutamine levels. Manuscript

V Samuelsson C, Howells T, Kumlien E, Enblad P, Hillered L,

Ronne-Engström E. Relationship between intracranial hemo-dynamics and microdialysis markers of energy metabolism and glutamate-glutamine turnover in subarachnoid hemor-rhage patients. Submitted

Reprints were made with permission of the publishers

Contents

Introduction...................................................................................................11 Acute brain injuries ..................................................................................11

Epidemiology.......................................................................................11 Neurointensive monitoring and care....................................................11

Brain blood flow and energy metabolism ................................................12 Ischemia...............................................................................................13

Seizures ....................................................................................................14 Glutamatergic neurotransmission.............................................................14

Glutamate uptake .................................................................................15 The glutamate-glutamine cycle............................................................16 The role of glutamate and GLT-1 in the injured brain ........................17

Microdialysis............................................................................................18 The technique ......................................................................................18 Recovery..............................................................................................19 What and where does microdialysis measure? ....................................20 Microdialysis to monitor brain energy metabolism.............................20

Aims of the Study .........................................................................................22

Materials and Methods..................................................................................23 Animals (Paper I & II) .............................................................................23

Iron-induced posttraumatic epilepsy model.........................................23 Microdialysis in animals......................................................................23 EEG .....................................................................................................24

Immunoblotting........................................................................................24 PBN analysis of blood..............................................................................25 Patients (Paper III-V) ...............................................................................25

Microdialysis in patients......................................................................26 Computerized multimodality monitoring ............................................26 Recording of clinical status and clinical course of events ...................27 Classification of microdialysis patterns...............................................27

Statistics ...................................................................................................28 Ethics........................................................................................................29

Results & Discussion ....................................................................................30 Time dependent protein alterations at the iron-induced epileptogenic lesion. .......................................................................................................30 Significance and reproducibility of the animal model .............................31 There is oxidation-induced aggregation of GLT-1 at the epileptogenic lesion. .......................................................................................................32 The interstitial glutamine/glutamate ratio is altered at the epileptogenic lesion ........................................................................................................34 Human interstitial brain glutamate and glutamine relate to the energy metabolism ...............................................................................................36 L/P ratios above 40 are associated with clinical criteria of ischemia.......40 Where was the probe, where should it have been and why? ....................41 Interstitial glutamine increase over time and glutamine surges may signal hard-working astrocytes.........................................................44

DINDs are associated with glutamine surges ......................................45 Glutamine surges appear when ICP is lowered ...................................46 Interstitial glutamine relate to the clinical admittance status...............47 Cessation of sedation does not explain the glutamine surges ..............48

Intracranial hemodynamics and microdialysis measurements .................48 Interstitial glutamine � what does it mean?..............................................51 Origin of low glutamine/glutamate ratios in animals and humans...........52 Methodological and statistical considerations..........................................53

Concluding remarks ......................................................................................55

Acknowledgements.......................................................................................56

References.....................................................................................................58

Abbreviations

AMPA: �-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid ATP: adenosine triphosphate BBB: blood-brain barrier BP: blood pressure CBF: cerebral blood flow CPP: cerebral perfusion pressure CSF: cerebrospinal fluid CT: computed tomography EAA: excitatory amino acids EAAT: excitatory amino acid transporter EEG: electroencephalogram GFAP: glial fibrillary acidic protein GLAST: glutamate and aspartate transport protein Gln: glutamine Glt: glutamate GLT-1: glutamate transporter-1 GS: glutamine synthetase HPLC: high performance liquid chromatography ICH: intracerebral hemorrhage ICP: intracranial pressure i.p.: intraperitoneally L/P: lactate/pyruvate mBP: mean blood pressure MD: microdialysis NICU: neurointensive care unit NMDA: N-methyl-D-aspartate PAG: phosphate activated glutaminase PBN: �-phenyl-tert-N-butyl nitrone PVDF: polyvinylidene difluoride RLS 85: Reaction Level Scale 85 SAH: subarachnoid hemorrhage SC: subcutaneous SD: standard deviation

SE: standard error of the mean SNAT 1-2: system A amino acid transporter SNAT 3: system N amino acid transporter TBI: traumatic brain injury TCA: tricarboxylic acid VSP: vasospasm WFNS: World Federation of Neurosurgical Surgeons

11

Introduction

Stroke and traumatic brain injury (TBI) are severe acute brain insults where secondary pathophysiological processes such as focal or global ischemia, altered energy metabolism, increased intracranial pressure (ICP) and excito-toxicity are commonly initiated. Neurointensive care aims at preventing loss of viable brain tissue due to these secondary processes which are believed to contribute significantly to morbidity and disability 3,34,170. Since TBI and stroke have high mortality rates, often cause long-lasting functional impair-ment in survivors and have great social and economic impact much can be gained if the brain damage progression can be hampered 58,195. To achieve this, profound knowledge about what goes on in the injured brain is required. This thesis is an attempt to improve the understanding of some aspects of the cerebral glutamate turnover, which appears to play a role in the pathophysi-ological processes following acute brain injury.

Acute brain injuries

Epidemiology Stroke, which includes ischemic cerebral infarct (85%), intracerebral hemor-rhage (ICH) (10%) and subarachnoid hemorrhage (SAH) (5%) annually strike approximately 20 000 persons in Sweden 146. National and interna-tional data indicate that around 250/100 000 are annually hospitalized due to TBI 6,23,177. Severe TBI, which often requires surgical intervention, accounts for 4-10% of all TBI cases 23,112,177.

Neurointensive monitoring and care Patients with acute brain injuries, typically TBI, SAH or ICH, are often ad-mitted to the neurointensive care unit (NICU) where aggressive monitoring and optimization of physiological variables aim at promoting recovery and improve outcome 34,170.

Since NICU patients often are sedated and/or have decreased conscious-ness they commonly require ventilatory support to assure adequate oxygena-tion. A major focus is placed on meticulous observation of the patients’ neu-

12

rological state and physiological parameters, including blood pressure (BP), ICP, cerebral perfusion pressure (CPP), SpO2, arterial gases, electrolytes and temperature, in order to detect preventable deteriorations and initiate diag-nostic procedures and treatment.

Brain blood flow and energy metabolism The brain has high energy demands: it represents 2% by weight, uses 20% of the oxygen and glucose supply, and receives 15% of the cardiac output in the resting adult body 169. Cerebral blood flow (CBF) normally remains rela-tively unaltered in the mean arterial blood pressure (mBP) range 60-150 mmHg, owing to an ability of the cerebral arterioles and precapillary sphinc-ters to vary their resistance. This mechanism is referred to as “cerebral auto-regulation” and ensures a sufficient supply of oxygen and nutrients to the brain without exposing it to unnecessary large blood volumes 68,129. How-ever, in traumatized or ischemic brain, the autoregulatory mechanism can be malfunctioning 63,82,99. As a consequence, CBF becomes increasingly blood pressure dependent, why hypovolemia and/or hypotension can have deleteri-ous effects 45. The CBF can also be reduced as a consequence of intracranial expansive processes (e.g. hematoma, edema or hydrocephalus) which elevate the ICP.

Glucose is the brain’s main energy substrate 164. Adenosine triphosphate (ATP) is produced almost exclusively by oxidative glucose metabolism, which is tightly linked to brain activity 132. In the resting state both astrocytes and neurons can take up glucose and completely oxidize it to CO2 and H2O, which yields an oxygen/glucose ratio that approaches six 39. During brain activity, astrocytes are however capable of a several-fold increase in glyco-lysis as a response to an increased energy demand. If there is a high glyco-lytic rate pyruvate and reduced nicotinamide adenine dinucleotides (NADH) will accumulate. To regenerate NAD+ (oxidized form of NADH) and restore the intracellular re-dox potential, pyruvate will be reduced to lactate by lac-tate dehydrogensase (LDH). Monocarboxylates (lactate and pyruvate) can be shuttled from astrocytes to neurons via the interstitial space 17 and neurons are capable of switching between glucose and monocarboxylates as oxida-tion substrates, depending on the relative availability in the interstitial space. This compartmentalization of anaerobic and aerobic glucose metabolism between astrocytes and neurons in situations of high cerebral metabolic de-mands has found increasing support during the last decade 90,108.

13

Ischemia Ischemia is a relative shortage of the blood supply, which results in inade-quate oxygenation and glucose delivery to the tissue. Ischemia can conse-quently occur as a result of decreased delivery of oxygenated blood or in-creased demand of oxygenated blood. During ischemia there is progressive ATP production failure in the cells with associated breakdown of mecha-nisms necessary for cellular function (functional ischemia) and survival (structural ischemia/infarct). Lack of oxygen at the mitochondrial level as well as mitochondrial inability to utilize oxygen, will lead to NADH accu-mulation, reduced cellular redox-state, and increased lactate production from pyruvate 29.

Stroke and TBI patients are at high risk of brain energy failure and ische-mia. Common systemic complications, such as hypoxia, hypotension, epilep-tic seizures and fever, can result in a mismatch between brain oxygen supply and demand. In addition, there may be spontaneous, post-operative or trau-matic lesions such as hematoma, contusion or edema, which further com-promise the blood delivery to adjacent regions 45.

Ischemia in SAH Following a SAH there are a number of specific, primary and secondary, causes of ischemic brain damage. When a cerebral artery ruptures the ICP can approach and exceed mBP levels, resulting in cerebral circulatory arrest and global brain ischemia with unconsciousness at the onset of the bleed 194. In the sub-acute phase following SAH there is a high risk of ischemia due to vasospasm (VSP), which is a seemingly non-purposeful constriction of the basal cerebral arteries. Using angiography, VSP is diagnosed in 60-70% of all SAH patients where narrowed segments of large caliber cerebral arteries can be visualized. However, only 20-40% of all SAH patients present with clinical symptoms of VSP, which include various degrees of neurological deterioration attributed to ischemia 14,100,101. The incidence of VSP is at high-est 4-10 days post SAH and risk factors include large amount of blood and bilirubin oxidation products in the subarachnoid space 69,138. Calcium an-tagonists, such as nimodipine which is believed to mediate its effects through neuroprotection and cerebral vasorelaxation, is routinely used as VSP prophylaxis in SAH patients 47. Delayed ischemic neurological deficits (DIND) following SAH are often attributed to VSP but may also be a conse-quence of interacting cerebral and/or systemic secondary complications 52. In case of DIND due to VSP, in the absence of structural (radiologically veri-fied) ischemia, hypertensive-, hypervolemic-, hemodilution-therapy is nor-mally initiated to improve cerebral blood flow and perfusion.

14

Seizures A seizure is the clinical manifestation of abnormal synchronized neuronal discharge. Epilepsy is diagnosed when seizures appear repeatedly and/or unprovoked. In partial epilepsy, where seizure activity originates from a definite region in the brain, there is frequently an identifiable cause in the past, e.g. TBI, status epilepticus, stroke, brain tumor or an infection 72. It appears as such insults can trigger cascades of neurochemical events, includ-ing generation of free radicals, leading to spontaneous seizures, referred to as remote symptomatic seizures, after a latency period 196. This latency period, corresponding to the epileptogenesis, normally ranges from months to years 30.

Seizures are common within the first week after TBI 60,190 and stroke 9,11,26,44,97,144, and are feared complications as they exert increased metabolic demands on an already vulnerable brain. Occurrence of acute seizures are prognostic of increased risk for developing chronic epilepsy 11,60,65,97. Pro-phylactic administration of anti-epileptic drugs reduces the incidence of early seizures, but has neither lead to increased survival nor to reduced inci-dence of remote symptomatic seizures 125,163. This indicates that early sei-zures are a manifestation of ongoing epileptogenic processes in the injured brain but that an early seizure activity per se does not contribute to epilepto-genesis.

The relative risk of developing epileptic seizures after stroke or TBI, compared to a non-stroke/TBI population, is between 1– 50 % depending on the severity and type of insult 30,60. Epilepsy is often a major component of disability and contributes to pre-mature death in patients who have survived TBI and stroke 1,36,172. Presence of early seizures, focal cortical damage, skull fractures, loss of consciousness, age, computed tomography (CT) - verified infarcts, cortical depositions of hemosiderin and blood in the sub-dural/subarachnoid space are risk factors for developing epilepsy following stroke and TBI 7-9,11,26,36,74.

Glutamatergic neurotransmission The main excitatory transmitter glutamate 54 binds to postsynaptic ligand-gated ion channels (NMDA-, AMPA- and kainate receptors), which leads to neuronal excitation through depolarization. The receptor activation is nor-mally transient but if it becomes excessive the target neuron will die. Such glutamate-induced cell death is called excitotoxicity 126 and involves pro-longed depolarization, ionic imbalance, ATP depletion and elevations of the intracellular free calcium concentration.

The average level of glutamate in the brain is ca �10 mmol/kg 147 but the transmitter is normally heavily compartmentalized in order to achieve a high

15

signal to noise ratio in the synaptic cleft. With microdialysis (MD) the basal interstitial glutamate concentration in human brain has been reported to be �15 μM 143,165 and most animal studies report concentrations in the low mi-cromolar range 5,103,119. There is however evidence that the NMDA receptor is activated by glutamate levels below 5 μM range 128 37 which, together with experimental evidence 75, implies that the resting synaptic glutamate concen-tration in normal brain is below this level.

Glutamate uptake High affinity, cell membrane bound, excitatory amino acid transporters (EAATs) are responsible for the removal of glutamate from the interstitial space and thereby aid in terminating synaptic transmission and prevent syn-aptic glutamate concentrations from rising to neurotoxic levels. Five differ-ent EAATs have been identified 4. The most abundant and functionally most important was in mouse originally named glial glutamate transporter 1, GLT-1 (human homologue is called EAAT-2) 148,154. It is localized predomi-nately on astroglial processes 42. GLT-1 accounts for some 90% of the glu-tamate uptake in the forebrain 71. The transport of glutamate into astrocytes requires energy and is driven by the co-transport of Na+ and counter-transport of K+ down their respective concentration gradients (Figure 1).

Figure 1. The astrocytic uptake of glutamate through GLT-1 is driven by flow of Na+ and K+ down their respective concentration gradients. The membrane potential is restored thanks to the action of Na+/K+ ATPase. Glt: glutamate, Gln: glutamine, GLT-1: glutamate transporter 1.

Upon dissipation of the electrochemical gradient across the cell membrane, e.g. during ischemia and energy failure, the GLT-1 mediated glutamate transport is impaired or even reversed with release of glutamate instead of uptake 182.

Glt K+

Na+

= GLT-1

ASTROCYTESYNAPTICCLEFT

K+

= Na+, K+ ATPase

16

The glutamate-glutamine cycle

After astrocytic uptake, glutamate is converted to glutamine by the ATP-dependent glia-specific enzyme glutamine synthetase (GS) which adds an amine group from ammonia to the glutamate carbon skeleton. The ammonia derives from the blood or from the brain’s metabolism. Astrocytes release glutamine back to the interstitial space via SNAT 3, a system N amino acid transporter 18,21. Neuronal uptake of glutamine occurs predominantly via system A amino acid transporters SNAT 1 and SNAT 2 21,46,87,173, but other transporters may also be involved in neuronal glutamine uptake 38. SNAT 3 can carry glutamine bi-directionally across the astrocytic cell membrane and glutamine efflux is favored by low extracellular pH 20. The intracellular glu-tamine concentration in both astrocytes and neurons normally exceeds the interstitial concentration 22,127, i.e. glutamine exits astrocytes down a concen-tration gradient and enters neurons against a concentration gradient. Inside the neuron glutamine is re-converted into glutamate by mitochondrial phos-phate activated glutaminase (PAG) in order to replenish the pre-synaptic glutamate transmitter pool 40. This efficient uptake and regeneration of glu-tamate and glutamine between neurons and astrocytes is known as the “glu-tamate-glutamine cycle” (Figure 2).

Figure 2. Schematic representation of the glutamate-glutamine cycle between neu-rons and astrocytes. Glt: glutamate, Gln: glutamine, GLT-1: glutamate transporter 1, SNAT 3: system N amino acid transporter, SNAT 1-2: system A amino acid trans-porter 1 and 2

Glt

Glt

Glt ---> Gln

Gln

Gln

= GLT-1= SNAT 3= SNAT 1-2

NH3

NH3

--->

ASTROCYTE

POSTSYNAPTICNEURON

PRESYNAPTICNEURON

17

The glutamate-glutamine cycle is the major metabolic pathway of both glu-tamate and glutamine in the brain. By labeling compounds radioactively it has been shown that the bulk of glutamate transmitter derives from gluta-mine 67,181 and neuronally released glutamate is the substrate in 80-90% of the astrocytic glutamine synthesis 89.

However, alternative pathways for glutamate turnover exist. Glutamate can enter the tricarboxylic acid (TCA) cycle in the astrocytes and be oxida-tively degraded instead of amidated 113,193. Astrocytes can also synthesize glutamate de novo by carboxylation of pyruvate, an anaplerosis estimated to account for 10-30% of the exported glutamine 76. High extracellular gluta-mate levels increase astrocytic glutamate oxidation and decrease pyruvate carboxylation in vitro 76. Thus the overall significance of the glutamate-glutamine cycle in the brain’s total glutamate turnover probably varies.

Reports on interstitial glutamine levels under normal and pathological condition range between 150-1200 μM 88,145,150,202. High glutamine levels have been considered neuroprotective 64,174 and the GS expression can in-crease in response to injury and high glutamate levels 139. Contrary to most other amino acids there is no significant concentration gradient for glutamine across the blood–brain barrier (BBB) 202 but the localization of glutamine transporters at the endothelial–astrocyte interface is such that it favors a net out-flux of glutamine from the brain 73.

Eighty percent of the total brain glucose consumption is used to drive the glutamate-glutamine cycle 109,167. Neuronal excitation, high extracellular glutamate levels and large astrocytic glutamate uptake are associated with increased astrocytic glucose utilization, primarily through augmented glyco-lysis 55,107,133,192. Glutamate uptake has correspondingly been shown to de-crease if the ATP generation is hampered 24.

High glutamate levels, low glutamine levels and slow glutamate-glutamine cycling have been described in epileptic brain 136 and low gluta-mine/glutamate ratios, measured with MD, were prognostic of bad outcome in children with TBI 145. This indicates that disturbed glutamate-glutamine cycling is of significance in the injured and/or metabolically compromised brain.

The role of glutamate and GLT-1 in the injured brain Excitotoxicity is a contributing factor to cell damage following acute brain insults 56 and elevated interstitial glutamate levels have been reported in TBI 25,123,189,203 as well as in SAH 43,53,134,158,162, commonly associated with high lactate/pyruvate (L/P) ratios. Several lines of evidence also support the role of excessive neuronal glutamate stimulation in epileptic seizures and epi-lepsy-induced brain damage 19,31,48,114,115,152,153,189.

The role of GLT-1 in the injured brain has received much attention. A transient GLT-1 decrease accompanied by decreased uptake of excitatory

18

amino acids (EAA) was observed after TBI 140 and decreased GLT-1 levels were associated with aggravated neuronal damage in models of TBI 141 and focal cerebral ischemia 142. Reversed glutamate transport through GLT-1 has been reported during ischemia and ATP-depletion 61,137. Altered levels of GLT-1 were found post-mortem in TBI patients and GLT-1 was absent in cerebral contusions despite high levels of the glia-specific protein glial fibril-lary acidic protein (GFAP), which was used as a marker for reactive astrocy-tosis 85. These results indicate that astrocytes may decrease their glutamate uptake ability in response to injury.

GLT-1 gene knock-out mice die of seizures 178 and reductions in GLT-1 encoding mRNA was observed in cortex, striatum and hippocampus of ge-netically epilepsy prone rats 2. Epileptogenic injections of iron in rat amyg-dala induced dynamic alterations in GLT-1 mRNA levels 184. However, there were no alterations in the levels of GLT-1 nor of GLT-1 mRNA in the hip-pocampus of fully kindled rats 3.

High GLT-1 activity has been correlated to neuronal survival following ischemic insults. Induction of the GLT-1 expression, which can be obtained pharmacologically by estrogen 130, �-lactam antibiotics 155, and glucocorti-coids 205 or by sublethal ischemic preconditioning 149,204, improves outcome and cell survival in ischemic models 35. Individuals with a common poly-morphism in the GLT-1 promoter region, which results in reduced GLT-1 gene transcription, have higher plasma glutamate levels and worse clinical course following thromboembolic stroke 110. However, since there is massive glutamate out-flux through the GLT-1 transporter upon dissipation of the membrane potential and energy failure 124, a high GLT-1 expression may also be harmful and contribute to neuronal death in case of overt ischemia 96.

Microdialysis

The technique With MD it is possible to monitor the neurochemical composition of the brain interstitial fluid. A thin (Ø�1mm) double lumen probe is inserted into the tissue and a physiological salt solution (the perfusate) is slowly pumped through the inlet tubing of the probe. At the dialysis-membrane-equipped tip of the probe the perfusate will equilibrate with the surrounding interstitial tissue fluid by diffusion. The perfusate, now called dialysate, will then con-tain a proportional concentration of the tissue fluid’s molecules 105,163 and will pass through an outlet tubing where it can be collected for analysis. To what extent a specific molecule crosses the probe membrane depends on its concentration gradient across the membrane as well as the size of the mem-brane pores. “Cut-off” is a term defined as the molecular weight at which

19

80% of the molecules do not pass the membrane. MD probes with cut-off 20 kDa are the most commonly used clinically. However probes exist in the cutoff range between 6 kDa to 3000 kDa.

Recovery The ratio between microdialysate concentration and interstitial fluid concen-tration is defined as the “relative recovery” and reflects to what extent the microdialysate concentration corresponds to the “true” interstitial concentra-tion 185.

Relative Recovery = Cmd / Cic = 1 – e - Ko A / F

Cmd = microdialysis concentration Cic = interstitial concentration Ko = average mass transfer coefficient (cut-off, temperature, diffusion char-acteristics) A = area of dialysis membrane F = flow rate of perfusate

Figure 3. Schematic graphs illustrating how the recovery in microdialysis is depend-ent on the flow rate of the dialysis fluid and the area of the dialysis membrane

The relative recovery of a substance, probe and flow rate at a specific tem-perature can be determined in vitro by inserting the probe into a fluid with a known concentration and analyzing the microdialysate. One must however be cautious to use in vitro recovery rates to calculate interstitial tissue con-centrations since the recovery in vivo is influenced by additional factors such as rate of diffusion in the interstitial fluid, active transport mechanisms, the

0

1

0 5 10 15flow rate of perfusate

rela

tive

reco

very

0

1

0 2 4 6 8area of dialysis membrane

rela

tive

reco

very

20

size of the interstitial room and alterations which inevitably will occur with time 15,16,105. Methods have been developed to estimate the recovery in vivo 92,116.

In the no-net flux method 106 the interstitial concentration of a certain sub-stance is determined by varying the concentration in the perfusate. The per-fusate concentration, at which there is no difference in the dialysate and per-fusate, i.e. where there is no net flux of the compound across the MD mem-brane, corresponds to the interstitial fluid concentration.

Another method is the extrapolation-to-zero-flow method where the tissue concentration is derived by varying the flow rate and extrapolating to the dialysate concentration where flow equals zero. Since a minimum amount of dialysate is required for analysis of compounds there is an inverse relation-ship between relative recovery and time-resolution at which the monitoring can take place.

Performing simultaneous MD in subcutaneous (SC) tissue and analyzing an endogenous reference compound, such as urea, can be used to monitor the performance of the intracerebral probe over time. Stable CNS-urea/SC-urea ratios indicate functioning probes 151.

What and where does microdialysis measure? The MD probe tip, inserted in brain tissue, will sample from the interstitial space, which is the fluid filled gap between the cells that normally make up �20% of the total brain volume. Several factors, including brain region, depth of insertion, neuronal activity as well as the presence or absence of brain pathology affect the characteristics of this space and consequently will influence the MD measurements 79.

Upon probe insertion there will invariably be a certain tissue trauma, which will be reflected in the dialysate. This tissue reaction appears to be transient and most metabolites will return to baseline within 1 hr after the insertion of a continuously perfused probe 59,78,117.

It must also be emphasized that although we, by using MD, can measure fluctuations in important brain metabolites, many biomarker signals are typi-cally complex and the technique can not alone answer the questions from where fluctuations in tissue concentration derive 79. Interpretation of MD data must consequently be done with caution. Several possibilities are com-monly plausible; in the case of glutamate an elevation may be due to neu-ronal synaptic release, carrier mediated release or overflow from non-synaptic sources, e.g. cell leakage or BBB breakdown.

Microdialysis to monitor brain energy metabolism MD was introduced as a neurochemical monitoring tool in neurointensive care in 1992 134. MD has since been used extensively to monitor the brain’s

21

energy metabolism since the interstitial levels of glucose, lactate and pyru-vate provide information about glucose and oxygen utilization and availabil-ity in the tissue. Currently the interstitial L/P ratio is regarded the most sensi-tive and specific MD marker of ischemia 13 and the L/P ratio has been shown to correlate with both the oxygen extraction ratio 53,83 and CBF 159.

The significance and specificity of different L/P ratio levels is however not established. Normal human MD L/P ratio levels range between 17-27 (based on mean reported values � 1 SD) 143,165. Given this, and the fact that that the interstitial L/P ratio is a reflection of the cellular redox state, L/P>40 has been suggested as a threshold value for “tissue energy crises” 188. Sup-port for this threshold are observations where L/P>40 occurred in human ischemic brain 81,94 and during permanent as well as transient cerebral blood flow reductions in experimental animal models 51.

Evidence however exist that cerebral energy crises can be a consequence of other phenomena than reduced blood flow since high L/P ratios have been described in the absence of both ischemia and hypoxia 81,188,191. Increasing attention has also recently been devoted to L/P ratio elevations with low pyruvate levels 77. These observations indicate that there may be a specific type of energy metabolic crisis that involves impaired glycolytic activity and/or shunting of glucose to other metabolic pathways 12,49.

22

Aims of the Study

The general aim of this thesis was to study glutamate turnover in acute brain injury. The specific aims were to investigate:

� GLT-1 levels in the iron-induced posttraumatic epilepsy model.

� if reductions in GLT-1 are associated with altered interstitial amino

acid levels and to study the effects of a free radical scavenger in the iron-induced posttraumatic epilepsy model.

� whether energy perturbation is generally associated with low glu-

tamine/glutamate ratios in injured human brain.

� microdialysis patterns of energy metabolism and glutamate turn-

over and their relation to clinical events in subarachnoid hemor-rhage patients receiving neurointensive care.

23

Materials and Methods

Animals (Paper I & II) Male Sprague Dawley rats (BK, Stockholm, Sweden), weighing 200-400 g were used. They were housed 2-3 / cage with free access to food and water. The animals were anaesthetized during all surgical procedures, MD, and at decapitation. Anesthesia was initiated using fluothane in oxygen and main-tained with 1-1.5% isoflurane in a 30/70 mix of O2 and N2O delivered through a nose cone or via an orotracheal tube. Body temperature was meas-ured rectally and maintained at 37.5 C° using a heating pad. During neuro-surgical and MD procedures the rats were fixated in a stereotaxic frame and the skin to be opened was infiltrated with 1% lidocaine. During MD rats received a catheter in the tail artery through which blood pressure was moni-tored and blood samples drawn. The rats arterial blood gases remained within the following ranges; pH 7.35-7.45, pCO2 4.5-6.9 kPa, pO2 11-19 kPa.

Iron-induced posttraumatic epilepsy model To create the epileptogenic lesion according to Willmore, a craniotomy was made on the left or the right side. Following opening of the dura 5 μl of 100 mM ferrous chloride solution was injected 5 mm laterally of the midline and 2 mm behind bregma 2 mm down the cortex at a rate of <1μl/min using a micropump (CMA 100/ Microdialysis AB, Stockholm, Sweden). The crani-otomy was closed with bone-wax and the skin was sutured. Before induction of the lesion, half of the animals in Paper II were intraperitoneally (i.p.) in-jected with the nitrone radical scavenger �-phenyl-tert-N-butyl nitrone (PBN) (100 mg/kg) (provided by Centaur Pharmaceuticals Inc., Sunnyvale, CA) dissolved in saline. Controls received i.p. saline.

Microdialysis in animals A MD probe (MAB 6.14.3, Microbiotech, Sweden) with a 15 kDa cut-off, 3 mm long membrane was inserted into the cortex adjacent to the iron-lesion or at the corresponding site in controls. Probes were perfused (2 μl/min) with

24

artificial cerebrospinal fluid (CSF; Na+ 148 mM, Ca2+ 1.2 mM, Mg2+ 0.9 mM, K+ 2.7 mM, Cl- 155 mM) and microdialysate samples (20 μl) were collected in 10-min fractions. The samples were analyzed for amino acids using high performance liquid chromatography (HPLC) with fluorescence detection (CMA/280 Microdialysis AB, Stockholm, Sweden). Twenty pri-mary amino acids were automatically pre-column o-phtaldialdehyd deriva-tised in a refrigerated CMA/200 autoinjector (Microdialysis AB, Stockholm, Sweden) and gradiently separated on a Nucleosil C18 reverse phase column as described previously 39. Glucose, lactate, and pyruvate were analyzed with enzymatic technique using CMA 600 Microdialysate Analyzer (Stockholm, Sweden). The basal value for each animal was calculated as the mean of three (for amino acids) or two (for metabolites) different fractions collected during the third hour of MD, allowing for wash out of any leakage of me-tabolites into the interstitial space caused by the insertion of the MD probe. The manufacturer report in vitro relative recovery for this probe and pump rate to be approximately 35% for lactate and pyruvate and 15% for gluta-mate.

EEG To study the reproducibility of the model, electroencephalograms (EEG) were done on some of the animals in Paper I in a three-channel montage using subcutaneous stainless needle electrodes and a digital acquisition sys-tem (‘Voyageur', Nicolet Inc. Madison, WI). The EEG recordings were ana-lyzed blindly by a co-authoring neurophysiologist without knowledge of whether the animal had been given an injection of ferrous chloride solution or CSF. Transients with an amplitude twice the background amplitude and with a duration of <70 ms were defined as spike potentials, while those with a duration of 70-200 ms were considered as sharp waves.

Immunoblotting For immunoblotting experiments in Paper I and II, the animals were decapi-tated, the brains were placed on ice and the tissue around the lesion was re-moved with a punch (Ø=4 mm), frozen immediately and stored at -75C°. Cell homogenate was prepared from the tissue in lysisbuffer containing pro-tease inhibitors. In Paper I the membrane fraction of the tissue was prepared using washing and ultracentrifugation steps as described previously 140. In Paper II whole cell extract was used. The samples were briefly sonicated and insoluble particles were spun down. 4μg (Paper I) and 12 μg (Paper II) total protein from each animal were prepared with a loading buffer containing sodium dodecyl sulfate (SDS) and �–mercaptoethanol. Electrophoreses was performed on polyacrylamide gels that were transferred to polyvinylidene

25

difluoride (PVDF) membranes (Bio-Rad laboratories, Hercules, CA). The membranes were incubated with anti-GLT-1 antibodies (Alpha Diagnostic International, San Antonio, Texas) followed by horseradish peroxidase-coupled anti-rabbit secondary antibodies. The protein bands were visualized using enhanced chemoluminiscence (Sigma Aldrich AB, Sweden). Im-munoblotting procedures were repeated using antibodies against GFAP (Sigma Aldrich AB, Sweden) (Paper I and II), glutamate and aspartate trans-port protein (GLAST) (Alpha Diagnostic International, San Antonio, Texas) (Paper I) and �-tubulin III (Biosite, Sweden) (Paper I) and visualized as above. The intensities of the bands were quantified using densiometric scan-ning and image software.

PBN analysis of blood Trunk blood was collected from the animals at decapitation and centrifuged at 5000 G for 5 min at 4C. The supernatant was collected and stored at -70C. Samples were kept dark at all times. PBN in plasma was analyzed using HPLC as described previously 111. Lowest detectable level was 2 μg/g. Twenty-four h after the cortical iron injection, at the time of sacrifice/MD; no PBN was detected in blood.

Patients (Paper III-V) Thirty-three patients (22 women and 11 men) with spontaneous SAH, admit-ted to Uppsala University Hospital NICU were studied. Mean age was 55.5 (standard deviation, SD 8.9) years. Patients were admitted between April 2003 and September 2005. All were believed to have a potential to survive. The median World Federation of Neurosurgical Surgeons (WFNS) grade 180 was 3.5 (range 1 to 5). The aneurysms were treated surgically (n=16) or with endovascular therapy (n=16). In one patient no aneurysm was found. Demo-graphic data is presented in Table 1. In the neurointensive care protocol, all SAH patients have bed rest and receive nimodipine. The aim is to keep the patients normovolemic and normotensive. Patients with decreased level of consciousness are intubated and artificially ventilated. All intubated patients are sedated with propofol. During the monitoring, all patients had a VD for ICP-monitoring and CSF drainage. We aim at keeping ICP below 20 and normally initiate CSF drainage at 15 mmHg. In case a delayed neurological deficit due to VSP is suspected and new ischemic lesions are absent on CT, the cerebral blood flow is optimized by increasing the blood volume and the blood pressure. The VSP treatment protocol includes maintaining a central venous pressure of 8-10 mmHg and systolic blood pressure above 140 mmHg, 0° head elevation, and administration of dextrane.

26

Microdialysis in patients A CMA/70 probe with 10 mm membrane length (CMA Microdialysis, Stockholm, Sweden) and 20 kDa cut-off were placed in frontal (n=30) or temporal (n=3) lobe cortex, either before or during the treatment of the aneu-rysm, via a separate burr hole in front of the VD (n=28) or through a bone flap (n=5). Care was taken to insert the probe in visually non-injured cortex. The mean duration from first sign of SAH to start of MD was 29.5 (SD 24) h and mean duration of MD sampling was 112 (SD 37) h. CT-scans showed no radiological signs of bleeding around the MD probe in any of the studied patients.

The probes were perfused with artificial CSF at a rate of 0.3 μL/min using a microinjection CMA/106 pump (CMA Microdialysis). Samples were taken hourly and analyzed bedside for lactate, pyruvate, and urea with enzymatic techniques using a CMA/600 Microdialysate Analyzer (CMA Microdialy-sis). Urea was monitored to control the probe performance 151. At least 3 h passed after insertion of the probe and start of sampling to allow for nor-malization of changes owing to probe insertion. The estimated relative re-covery using this MD set-up is �70% 84.

The CMA/600 Microdialysate Analyzer was automatically calibrated when started as well as every sixth hour using standard calibration solutions from the manufacturer. Quality controls at two different concentrations for each substance were performed every weekday. Imprecision values for be-tween-assay coefficient of variation was <10% for the low and <5% for the high concentration control samples.

Samples were analyzed for glutamate and glutamine with HPLC with fluorescence detection. Briefly, the amino acids were automatically derivat-ized with o-phthaldialdehyde in a CMA/200 (CMA Microdialysis) refriger-ated microsampler and gradient separated on a Nucleosil C18 column 5 μm, 60 mm x 4 mm (CMA Microdialysis). The mobile phase gradient consisted of 0.1 M sodium acetate buffer pH 6.7 with 2.5% tetrahydrofurane and methanol concentration increasing from 5% to 97.5%.

Quality control measurements of the HPLC system using blank water samples, as well as standardized calibration solutions mimicking those of the human samples were run daily. In addition, internal control samples were run together with each patient series. Imprecision values were 3% to 5% for within-assay coefficient of variation and 5% to 9% for between-assay coeffi-cient of variation for glutamate and glutamine.

Computerized multimodality monitoring A networked computerized multimodality monitoring system at the Uppsala NICU connected to a central server, is used to collect and store physiological patient data such as ICP, mBP, CPP, temperature and SpO2 82. One value per

27

minute is displayed and saved for each channel and patient. The data collec-tion is sometimes interrupted when the patient is taken to the operating thea-tre or during radiological examinations. The data was manually validated and the software was used to identify extreme values 82. Negative ICP values were either removed or set to zero. In total, 3617 hours at the NICU were monitored with MD and the computerized multimodality monitoring system. For correlation analyses each hourly MD sample measurement was matched to the mean ICP, MAP and CPP during the hour when the sample was col-lected, taking into account the dead space in the MD tubing which corre-sponded to a lag phase of 17 minutes. The dialysis fluid in a sample conse-quently reflected the brain interstitial milieu during the period -77 minutes to -17 minutes from the stamped collection time.

Recording of clinical status and clinical course of events Patients were grouped depending on WFNS grade at admittance and WFNS grade at the end of monitoring. WFNS grade 1-3 was regarded as good and WFNS grade 4-5 was regarded as bad clinical status. The patients' con-sciousness level and peripheral neurological status were recorded at 1-hour intervals by the NICU staff throughout the monitoring. Evaluation of the consciousness level was done using Reaction Level Scale 85 (RLS 85) 171. A clinical neurological deterioration was defined as an instant or period during which the patient consciousness level decreased (� a one-step increase in the RLS-score) or a new focal neurological deficit appeared. All clinical neuro-logical deteriorations that appeared after initiation of MD were noted. In order to be classified as an event the deterioration should have resulted in an action from the neurosurgeon in charge (e.g. a new CT-scan). Patient charts were retrospectively scrutinized for information regarding CSF drainage, fever-episodes, cardiopulmonary problems, surgical procedures, extubations and radiological examinations. The deteriorations were suspected to be a delayed ischemic neurological deterioration (DIND) unless there were other more likely explanations (e.g. hydrocephalus, fever, hypoxia, and electrolyte disturbance). Brain CT-scans of the patients obtained during and after the monitoring time where retrospectively studied to obtain information about ischemia and bleedings as well as the exact position of the MD probe and its localization to the parent artery of the aneurysm.

Classification of microdialysis patterns For each patient, the interstitial concentration of lactate, pyruvate, glutamate, glutamine, L/P ratio, and glutamine/glutamate were plotted over time and analyzed for patterns. Periods where the L/P ratio reached above 40, indicat-ing a pattern of “ischemia” and/or “energy crisis” 188, were identified. In Paper III at least two consecutive L/P ratios above 40 were required to be

28

considered an “energy crises event”. In Paper IV at least five hours with L/P>40 during a period of 10 hours were required to be classified as an “ischemic pattern”.

In Paper II, the concentration curves for glutamate and glutamine during energy failure periods were examined to determine if interstitial glutamine and glutamate increased or decreased during the period when L/P ratio reached above 40. A “glutamate increase” was when the glutamate concen-tration was higher in all the “high L/P ratio samples” as compared with the preceding glutamate baseline concentration. In addition, the interstitial glu-tamate levels should reach above the normal upper reference limit (32 μM143) to be classified as an increase. For glutamine, it is currently less clear what to regard as a normal concentration. We defined a “glutamine increase” when the glutamine concentration was higher in all the high L/P ratio sam-ples as compared with the preceding glutamine baseline concentration and/or there was a clear upward slope of the glutamine curve during the high L/P ratio period. A “glutamine decrease” was when the glutamine concentration was lower in all high L/P ratio samples as compared to the preceding gluta-mine baseline concentration and/or there was a clear downward slope of the glutamine curve during the high L/P ratio period. A pattern where no obvi-ous pattern was discerned, for example in the case where there were both increasing and decreasing interstitial glutamine concentrations during the high L/P ratio period, was classified as “varying”.

In Paper IV all accelerated increases in interstitial glutamine, “glutamine surges”, were identified. A “glutamine surge” was a period of at least 12 h where the interstitial glutamine concentration increased with at least 150 μM per 12 h.

Statistics Statistical analyses and graphical presentations were performed using Statis-tica 7.1. for Windows (StatSoft Inc., Tulsa, OK, USA). Results have been regarded significant if P<0.05. Correlation coefficients above 0.20 were considered to be of potential significance. Unless stated otherwise figures in brackets represent mean values � SD. Parametric and non-parametric statis-tics on raw and transformed data were used to describe, analyze, and com-pare MD data.

In Paper I and II, unpaired t-test on raw data was used to compare protein amounts on immunoblots as well as interstitial levels of amino acids and energy metabolites, in different groups.

Raw and logged MD data together with raw pressure data were used for correlation analyses in Paper III and V. Missing data was pair-wise deleted. To test differences in Pearson correlation coefficients, Fisher's z transforma-tions were conducted.

29

In Paper III, Wilcoxon matched pairs test was used to compare glutamine levels at start and end of monitoring. Kruskal–Wallis analysis of variance with post hoc testing and multiple comparisons of mean ranks was used to compare glutamine and glutamate during high L/P ratio periods, grouped according to pyruvate level.

In Paper IV, WFNS grade groups were compared with respect to gluta-mine levels at different time points using unpaired t-test. Fisher’s exact test was used to determine if there was an association between glutamine surges and extubations.

In Paper V, monitoring hours with ICP>10 mmHg were compared to hours with ICP<10 mmHg using unpaired t-test and Mann-Whitney U-test.

Ethics The local ethics committee for animal and human research approved all stud-ies. Patients or their relatives had given informed consent to participation.

30

Results & Discussion

Time dependent protein alterations at the iron-induced epileptogenic lesion. Both GLT-1 and GLAST are almost exclusively found on astrocytes, with GLT-1 found mainly in cortex and hippocampus, and GLAST found mainly in cerebellum 41. Paper I was performed in order to investigate GLT-1 and GLAST in the epileptogenic region at different time points following the cortical iron injection. Nine epileptic animals and six control animals were used.

Using immunoblotting techniques the level of GLT-1 (~70 kDa) appeared to be decreased by 38% at day 1 and by 70% at day 5 in the iron-lesioned animals compared to controls (Figure 4). After 3 months, there was no sig-nificant difference between the groups, but those results were inconclusive since the variation in GLT-1 levels was great between the animals. GLAST (67 kDa) and �-tubulin III (50 kDa) levels were unchanged at all times after the cortical iron injection.

Figure 4. Immunoblots of control (C) and iron-lesioned (E) animals 5 days after induction of the lesion. Cortex from iron-lesioned animals clearly have decreased intensity of the band corresponding to GLT-1 (~70 kDa), compared with control animals. An extra band (*) was observed at ~ 45 kDa for GLT-1.

31

The amount of GFAP increased with time compared to controls (Figure 5). The progressive and delayed astrocytic reactivity, indicated by GFAP in-crease at the iron-lesion, may be of importance in the development of epilep-togenic lesions. This is supported by studies in patients with posttraumatic seizures, where hemosiderin deposits surrounded by gliosis were identified 7,98. Gliosis is also a common finding in resected epileptic tissue 57.

Figure 5. GFAP levels at different time-points were compared by measuring the light absorbance using densiometric scanning of the immunoblots. The relative ab-sorbance shows GFAP levels in iron-lesioned animals relative to control animals. Mean relative absorbency (� SD) at day 1, 5 and 90 after induction of the iron-lesion is presented. There was a progressive relative increase of the amount of GFAP in iron-lesioned animals, compared to controls.

Significance and reproducibility of the animal model Bleeding within the cerebral tissue is associated with the development of posttraumatic and post-stroke epilepsy. Extravasation of blood leads to he-molysis and deposition of heme-containing compounds in the tissue. This can initiate a sequence of redox-reactions and free radical generation leading to peroxidation of lipids and proteins resulting in cell death and the forma-tion of an epileptogenic focus 199. Histologically, iron-associated epilepto-genic foci in humans and animals are characterized by gliosis, neuronal loss, neuronal deformity and accumulation of iron in macrophage-like cells 10,27,70,80,151.

All animals that received a cortical iron injection in Paper I displayed epi-leptiform activity at 3 months, with frequent spike/sharp wave activity (Fig-ure 6). Control rats had no epileptiform EEG activity. This confirms the re-producibility of the model. In other studies more than 90% of the animals present epileptiform activity 121,153 and epileptogenesis occurred in stages for up to 12 months following the cortical iron injection 121. This resembles hu-

32

man posttraumatic epilepsy where epileptiform activity usually arise within the first year following the brain injury 7.

Figure 6. An EEG recording from a rat, three months after a cortical iron injection. Frequent spike and sharp wave activity is present. T3 was placed over the left and T4 over the right temperoparietal region. Fz was frontal, in the midline

There is oxidation-induced aggregation of GLT-1 at the epileptogenic lesion. In Paper II we extended our investigation of astrocytic glutamate uptake in the posttraumatic iron-induced epilepsy model. The levels of GLT-1 as well as of interstitial amino acids and metabolites were measured using im-munoblotting and MD 24 h after a cortical iron injection. Since formation of free radicals is involved in epileptogenesis following stroke and TBI, the effect of a radical scavenger was studied.

Half the animals were given PBN intraperitoneally in a dose that had proven neuroprotective in other brain injury models 28. Thirty-eight animals were studied and there were four study groups: iron-lesion (F); iron-lesion + PBN (FP); no lesion (C); no lesion + PBN (CP).

There were decreased amounts of GLT-1 monomer (~70 kDa) at the iron-lesion compared to controls. High-molecular weight GLT-1 species were present at the iron-lesion but absent in controls. Iron-lesioned animals pre-treated with PBN had more of the 70 kDa band and less of the high-molecular weight bands compared to the iron-lesioned animals not receiving PBN (Figure 7). The total GLT-1 immunoreactivity, the summed intensities of all three bands, was similar in the F group and the FP group. The control groups C and CP had similar GLT-1 immunoreactivity. GFAP did not differ between the four groups, which is in line with the results from Paper I where

33

increased GFAP immunoreactivity at the iron-lesion appeared at 5 days and 3 months but not at day 1 (Figure 5).

Figure 7. GLT-1 immunoreactivity 24 h after a cortical iron injection and the effects of PBN. Immunoblots using antibodies against GLT-1 and GFAP on cell homoge-nates from iron-lesioned cortex. Each column represents one individual animal. F = homogenate from iron-lesioned cortex, FP = homogenate from iron-lesioned cortex from animals pre-treated with PBN, C = control homogenate from cortex where artificial CSF was injected instead of iron. The GLT-1 70 kDa band was completely absent on one FP animal at the edge of this blot (indicated with •), probably due to technical reasons. Repeat of the immunoblotting procedure for that animal revealed a band pattern similar to the other FP animals, including a 70 kDa band.

Since PBN attenuated the appearance of high molecular weight species of GLT-1, the results indicate that there is oxidation-induced aggregation of GLT-1 at the lesion site. GLT-1 transporters are known to occur in pairs on the cell membrane and steric interaction between two transporters is required for glutamate uptake. In the presence of an oxidant, a GLT-1 pair can bind to each other covalently, impairing their motility. On immunoblots, aggregated transporters will appear as bands with molecular weight about two or three times that of the ~70 kDa monomer 71. The two larger bands (150 kD and 220 kD) seen in iron-lesioned animals (Figure 7) presumably represent such multimers of GLT-1, induced by the presence of iron in cortex.

Aggregation of GLT-1, as demonstrated in Paper II, is also the most likely explanation behind the decreased levels of GLT-1 monomer reported at day 1 and day 5 in Paper I. However, after having confirmed the GLT-1

34

antibody specificity on normal brain homogenate, the high molecular weight part of the PVDF membrane was cut off in subsequent Paper I experiments (in order to conserve antibodies). In hindsight this was obviously erroneous and explains why only GLT-1 bands below 140 kDa were detected in in the iron-lesioned animals in Paper I.

Storage of brain tissue can induce oxidation-mediated aggregation of glu-tamate transporters 71. The punched-out brain tissue was stored in -75°C for 1-3 months in Paper I and for 6 months in Paper II, before the brain ho-mogenate was prepared. It is possible that the oxidation/aggregation oc-curred during this storage. However, the striking effects of PBN indicate that aggregation occurred during the first 24 h following the cortical iron injec-tion and argues in favor of this will now be outlined: Pharmacokinetic stud-ies show that peak PBN brain concentration following peripheral administra-tion appear after 30 minutes, PBN has a biological half-life of 134 minutes, and the brain concentration of PBN is similar to, or below, the plasma con-centration 32,33. We could not detect PBN in plasma 24 h after our i.p. injec-tion. Therefore we assume that no PBN was present in the brain when the tissue was collected and we argue that the major iron-induced oxidation of GLT-1 and attenuation of oxidation by PBN – resulting in the different band patterns in Figure 7 – occurred in vivo and not during storage.

The interstitial glutamine/glutamate ratio is altered at the epileptogenic lesion Interstitial glutamate concentrations were higher and glutamine concentra-tions were lower at the iron-lesion as compared to normal tissue (Figure 8). There was a trend, but no statistically significant difference, towards normal-ized glutamine/glutamate ratios in PBN pre-treated iron-lesioned animals.

GLT-1 mediated glutamate transport is considered the most important mechanism to maintain low interstitial levels of glutamate in cortex 148. The low glutamine/glutamate ratios at the iron-induced epileptogenic lesions may be a consequence of decreased glial glutamate uptake. Other possible expla-nations however exist. As outlined above, energy failure with high L/P ratios and collapse of the electrochemical gradient across the cell membrane result in astrocytic glutamate release instead of uptake 43,134,182. The L/P ratios in the iron-lesioned animals (mean 26 ± 6.3) were however similar to the con-trol group C (mean 27 ± 1.7), which indicate that the probe sampled from viable tissue without energy failure. It is possible that the glutamate increase at the lesion represent unspecific leakage through a disrupted BBB, in line with what has been described in human TBI where increased EAA levels were associated with increase in structural amino acids, e.g. threonine 25. There are steep concentration gradients across the BBB for most amino ac-

35

ids, with concentrations being far higher in the blood. The reported blood plasma concentration of glutamate and of threonine 202 is approximately 60 times higher than the respective concentration in microdialysate obtained from group C in our study. Both glutamate and threonine were increased in the iron-lesioned brains, but the relative increase in glutamate is more than twice the relative increase in threonine. This suggests a mechanism other than unspecific leakage behind the elevated glutamate at the iron-lesion.

glutamate glutamine glutamine/glutamate0.50.60.70.80.91.0

2.0

3.0

4.0

5.06.07.08.09.0

10.0

20.0

30.0

40.0

50.060.070.080.0

μM o

r no

unit

F FP C CP

Figure 8. Mean interstitial microdialysis concentrations (� SD) of glutamate, gluta-mine and the glutamine/glutamate ratio in cortical tissue 24 h after a cortical iron injection. F (n=6, iron-lesion); FP (n=6, iron-lesion + PBN.); C (n=3, no lesion); CP (n=3, no lesion + PBN).

Strong oxidants, such as Fe3+, can alter protein conformation, leading to distortion of secondary and tertiary structure, increased aggregation and fragmentation. Such alterations commonly make a protein susceptible to proteolysis and diminution of normal function. It has been shown that lipid oxidation products can impair astrocytic glutamate transport 93 including GLT-1 activity 183. Oxidative stress further has a role in epileptogenesis and antioxidants have proven to efficiently prevent brain damage and the evolu-tion of epileptiform activity 86,95,104,131,186,197,198.

We have shown that the major protein responsible for clearance of synap-tic glutamate is structurally impaired and that this is associated with elevated interstitial glutamate levels and decreased glutamine levels, in the iron-induced posttraumatic epilepsy model. Our results suggest, on a molecular

36

basis, one possible epileptogenic mechanism in bleeding-induced epilepto-genic lesions.

Human interstitial brain glutamate and glutamine relate to the energy metabolism Considering the energy demanding cycling of glutamate and glutamine be-tween neurons and astrocytes we hypothesized that the interstitial gluta-mine/glutamate ratio may be an indicator of the brain’s energy status. In Paper III we examined if energy perturbation, evaluated mainly by the L/P ratio, is associated with low glutamine/glutamate ratios in the brain. MD data from visually non-injured cortex in patients with spontaneous SAH, during their first week of neurointensive care, was used.

We found that glutamate correlate positively with the L/P ratio (r = 0.68) and to lactate (r = 0.65). The glutamine/glutamate ratio correlated negatively with the L/P ratio (r = -0.66). All these relationships were of log–log nature (Figure 9). There was no correlation between interstitial glutamine and L/P ratio and no relation between glutamate and glutamine.

The observed correlation between glutamate and L/P ratio is in line with previous observations 135,157,158. Noteworthy is also that high glutamate and low glutamine/glutamate ratios predominantly occurred when the L/P ratio exceeded 40 (Figure 9, top), findings that further support L/P ratio 40 as a threshold for “energy crisis” 188 at which glutamate levels are high 135.

There was a significant correlation between interstitial levels of glutamine and pyruvate (r = 0.55) and this relationship appeared linear (Figure 10A). In 22 out of 33 patients, the glutamine curve and pyruvate curve had similar shapes (Figure 10B).

Since transiently reduced glutamine production associated with energy perturbation could have been missed in the gross correlation analysis where the majority of the MD samples represented normal L/P-ratios with low glu-tamate levels indicating an efficient astrocytic glutamate uptake, we specifi-cally identified “energy crisis”-periods where L/P ratio exceeded 40. For these periods we determined whether interstitial glutamine and glutamate increased or decreased as compared to the preceding baseline concentration.

37

Figure 9. Scatterplots of raw (top) and logged (bottom) concentration in MD sam-ples collected from 33 SAH patients at the NICU. (A) Lactate/pyruvate vs glutamate. (B) Lactate/pyruvate vs glutamine/glutamate.

Figure 10. Scatterplot of MD glutamine vs. pyruvate in 33 SAH patients at the NICU. Pearson linear correlation coefficient of raw data in graph is 0.52, P<0.01. (B) Glutamine and pyruvate concentrations over time in one patient to exemplify the close relation between the two compounds.

0 500 1000 1500 2000 2500

glutamine (�M)

-100

0

100

200

300

400

500

pyru

vate

(�M

)

0 20 40 60 80 100 120 140

hours from onset of SAH

50

100

150

200

250

300

350

pyr

uvat

e (μ

M)

0

500

1000

1500 g

luta

min

e (μ

M)

A B

0 50 100 150 200 250 300 350

glutamate (�M)

0

30

60

90

120

150

lact

ate

/ pyr

uvat

e

0 1000 2000 3000 4000

glutamine / glutamate

0

30

60

90

120

150

lact

ate

/ pyr

uvat

e

0 2 4 6 8

log glutamine / glutamate

1

2

3

4

5

log

lact

ate

/ pyr

uvat

e

0 2 4 6

log glutamate

1

2

3

4

5

log

lact

ate

/ pyr

uvat

eA B

38

Thirteen periods with “energy crisis” (median duration 20 h, range 2 to 120 h, total duration 404 h) were identified. The median L/P ratio of these 404 samples was 53 (range 40 to 141) and median glutamate level was 110 μM (range 33 to 320 μM). In four out of 13 periods with L/P>40, the interstitial glutamine levels decreased (Figure 11A). In seven out of 13 periods with L/P>40, there was a varying response involving both increasing and decreas-ing glutamine concentration. In two out of 13 periods where L/P>40, the interstitial glutamine concentration increased, concomitantly to increasing glutamate, lactate, and pyruvate (Figure 11B).

Given the strong correlation between glutamate and L/P ratio it was not surprising to see that high L/P ratio periods always were associated with coincident sharp increases in interstitial glutamate. During overt ischemic conditions it has been shown that extracellular glutamate accumulation mainly is derived from reversed astrocytic glutamate transport due to failing astrocytic membrane potentials 166,176. If glutamate leaves the astrocytes, intracellular glutamate concentration will fall and there will be less substrate for GS. A decreased glutamine synthesis would presumably be reflected in the interstitial space since glutamine normally leaves the astrocytes by diffu-sion.

Figure 11. Examples of MD measurements from two patients to illustrate decreasing vs. increasing interstitial glutamine concentration during an energy crisis period when L/P>40. (A) L/P ratio and glutamate increase sharply at hours 107 to 109. Interstitial glutamine and pyruvate decrease during these 3 h compared with the preceding baseline readings. (B) L/P ratio exceeds 40 during hours 64 to 85. Gluta-mate, glutamine, and pyruvate increase during this period compared with the preced-ing baseline readings.

90 95 100 105 110 115 120 125

hours from SAH

0

20

40

60

80

100

120

L/P

glut

amat

e (μ

M)

0

100

200

300

400

500

600

700

800

900

1000

A

50 60 70 80 90 100 110 120 130

hours from SAH

0

20

40

60

80

100

120

200

400

600

800

1000

1200

1400

1600

1800

2000

2200

2400

2600

pyr

uvat

e (μ

M)

g

luta

min

e (μ

M)B

39

However, when analyzing glutamine during the 13 energy perturbation peri-ods, no definitive pattern emerged; interstitial glutamine concentration some-times increased and sometimes decreased. Only when pyruvate was consid-ered it was possible to predict whether glutamine and the gluta-mine/glutamate ratio would increase or decrease during the high L/P ratio period. The relationship between interstitial glutamine and pyruvate, present in the entire material (Figure 10A) as well as in the sub-group of high L/P ratio samples (Figure 12), is a novel observation that may enhance the un-derstanding of the glutamate-glutamine cycle energetics and metabolic dys-function following brain injury.

Figure 12. Glutamine (Gln) concentration and glutamine/glutamate (Gln/Glt) ratios in MD samples where interstitial L/P>40 (n=404) grouped depending on whether the pyruvate concentration was low (<119 �M, n=214), normal (119-213 �M, n=169) or high (>213 �M, n=21) 143. Plots represent median values with the 25%-75% range. The glutamine concentration differed significantly between the three groups (P<0.01). Glutamine/glutamate ratios were lower in the low pyruvate group com-pared to the normal pyruvate group and the high pyruvate group (P<0.05). The difference in glutamine/glutamate ratio between the high pyruvate group and normal pyruvate group did not reach statistical significance.

low normal high

pyruvate concentration

200

400

600

800

1000

1200

1400

1600

Gln

con

cent

ratio

n (μ

M)

2

4

6

8

10

12

14

16

Gln

/ G

lt ra

tio

40

L/P ratios above 40 are associated with clinical criteria of ischemia Patient demographic data, aneurysm localization, description of clinical events and MD patters are outlined in Table 1 and Table 2. In Paper IV an “ischemic MD pattern” was defined as at least 5 hours with L/P>40 during a 10 hour period. There were 12 ischemic MD periods in the material (mean duration 28 h, range 7-120 h).

Six ischemic MD patterns appeared as an L/P ratio increase from a lower baseline concentration (range 10-30 μM) and preceding a DIND (Figure 13, Table 1 and 2). The mean time from the first L/P ratio measurement above 40 to the DIND was 16.7 h (range 5-24 h, SD 8.7 h).

Five patients had ischemic MD pattern at onset of monitoring and there were consequently no baseline measurements. In four out of these five pa-tients, MD was initiated following craniotomy, aneurysm clipping and hema-toma-evacuation. The remaining one had the MD probe and a VD inserted due to a rapid neurological deterioration (Table 1 and Table 2). All patients with ischemic MD pattern from monitoring onset developed CT verified infarcts.

In one patient an ischemic MD pattern occurred as an L/P ratio increase from a low baseline during a radiological examination where the anesthesia was complicated. No infarcts were present on subsequent CT-scans of this patient’s brain (Table 1 and Table 2).

These observations indicate that L/P>40 is a severely pathological MD pattern that was associated with functional and/or structural ischemia in 11 out of 12 cases. L/P ratio elevations above 40 from a lower baseline also appear to be an early and specific warning pattern of an upcoming ischemic neurological deterioration since this pattern, 6 out of 7 times, was followed by a DIND.

In our experience MD baseline levels are hard to define, especially during ongoing monitoring. If the L/P ratio is to be used in clinical decision-making there are practical and didactic advantages to use an absolute threshold value instead of a relative increase from baseline. In the present study an absolute threshold of 40 was used. Using this value to predict DINDs there would have been few false positive cases. However L/P ratio measurements above 40 preceded only 40% (6 out of 15) of all DINDs in our material. This low sensitivity may be a consequence of the probe not being close to the ische-mic region and/or the fact that L/P>40 presumably only appear in tissue with severely perturbed energy metabolism.

41

Figure 13. MD patterns in patient #4 who experienced a DIND during monitoring. Arrows indicate DIND onset. The DIND is preceded by an ischemic MD pattern where L/P ratio exceeds 40. Glutamate levels are high. See Table 1 and 2 for more details.

Where was the probe, where should it have been and why? From Table 1 it is apparent that the relationship between the probe, focal lesions and the vascular territory of arteries varied amongst the patients. This was one major concern when initialising the analysis of the MD patterns and their relationship to clinical events.

In 8 patients the MD probe was localized in a territory supplied by the parent artery of the ruptured aneurysm. In 12 patients the MD probe was localized in a territory that may have been supplied by the parent artery of the ruptured aneurysm. In 12 patients the probe was localized in tissue that was not supplied by the parent artery of the ruptured aneurysm. There were six cases where an ischemic MD pattern preceded a DIND. In four out of these six cases the MD probe sampled from tissue that was not supplied by the affected artery.

20 30 40 50 60 70 80 90 100 110

hours after SAH

0

200

400

600

800

1000

1200gl

utam

ine

glut

amat

e (μ

M)

0

50

100

150

200

250

300

350

400

450

pyru

vate

(μM

)

0

10

20

30

40

50

60

L/P

42

In previous attempts to predict DIND using MD, care has commonly been taken to insert the probe in the tissue most likely to be affected by VSP 160,168. Skjoth-Rasmussen and co-workers showed that DINDs were preceded by 20% increases in L/P ratio from baseline with a positive predictive value of 85% and peak L/P ratio occurred 4-50 h (mean 23) before DIND onset 168. We adhered to our routine clinical practice where the most convenient time for probe insertion commonly is in conjunction with emergency ventricu-lostomy most often without a priori knowledge of the aneurysm localization. Nevertheless, in 6 out of 7 (86%) cases where L/P ratio increased above 40 from a low baseline this was followed by a DIND. On average, the first measurement of L/P>40 occurred 16.7 h (range 5-24 h) prior to the DIND. This yields a specificity and a time-window similar to that reported by Skjoth-Rasmussen et al. Noteworthy is also that in 4 out of 6 DINDs that were preceded by an ischemic MD pattern in our cohort, the probe was out-side the territory supplied by the aneurysmatic artery. This indicates that perhaps little is gained by individualizing the MD probe placement in SAH patients and a more standardized, global, placement of the probe may be simpler and yield equally specific information.

Table 1. (see page 43) Patient demographics, aneurysm localization, treatment re-gimes, MD information, occurrence of DIND and associated radiological findings and MD patterns in the 33 studied SAH patients. PC = posterior circulation, ICA = internal carotid artery, MCA = medial cerebral artery, ACA = anterior communicat-ing artery, E = embolization, C = clipping, VD = with ventricular drain, BF = via bone flap, R.fro = right frontal lobe, L.fro = left frontal lobe, R.tem = right temporal lobe, L.tem = left temporal lobe, Y = yes, N = no, M = maybe, D = glutamine surge occurs during DIND, A = glutamine surge occurs after DIND, I = new ischemic lesion on CT, H = new hemorrhagic lesion on CT, 0 = no new lesion on CT, V = patient received VSP treatment, LP = L/P>40 occurs as an increase from baseline preceding a DIND, LP* = L/P>40 occurs from the beginning of monitoring, LP** = L/P>40 occurrs as an increase from a lower baseline during a complicated anesthesia and not associated with a DIND.

43

Patie

nt #

sex

age

aneu

rysm

treat

men

t

prob

e in

serti

on

prob

e lo

caliz

atio

n

sam

plin

g ve

ssel

te

rrito

ry

DIN

D

isch

emic

MD

pat

-te

rn

glut

amin

e su

rge

at

DIN

D

CT

resu

lt

VSP

trea

tmen

t

1 F 73 PC E VD R.fro N Y D 0 V 2 M 69 L. ICA E VD R.fro N Y A I 3 M 67 ACA E VD R.fro M Y LP A 0 V 4 M 63 L. ICA E VD R.fro N Y LP 0 5 M 60 ACA C VD L.fro M Y I 6 M 58 ACA C VD R.fro N Y LP A IH 7 F 56 L. MCA E VD R.fro N Y LP D 0 8 F 56 ACA E VD R.fro M Y D 0 V 9 M 53 ACA E VD R.fro Y Y LP D IH 10 F 52 R. ICA E VD R.fro N Y LP D 0 V 11 F 50 L. MCA C VD R.fro N Y D 0 V 12 M 47 PC E VD R.fro N Y A I 13 F 46 ACA E VD R.fro M Y D 0 V 14 F 44 L. MCA C VD R.fro Y Y LP** 0 15 F 69 ACA C BF L.tem N 16 M 64 ACA C VD R.fro M 17 F 64 ACA C VD L.fro Y Y D 0 V 18 M 62 ACA C VD R.fro M 19 F 59 ACA C VD R.fro M 20 F 55 R. MCA C VD R.fro Y 21 F 56 R. MCA C BF R.tem Y 22 F 53 R. MCA C BF R.fro Y LP* 23 F 51 ACA E VD R.fro M 24 F 50 L. MCA C BF L.tem Y LP* 25 M 50 L. MCA C BF L.fro Y LP* 26 F 50 none VD R.fro - 27 F 48 L. MCA C VD L.fro N LP* 28 F 46 R. ICA E VD R.fro N 29 M 42 L. ICA E VD R.fro M 30 F 35 L. MCA C VD R.fro N 31 F 54 ACA E VD R.fro M 32 F 70 ACA E VD R.fro M LP* 33 F 59 L. MCA E VD R.fro M

44

Table 2. Description of the 15 DINDs that occurred in the 33 patients, DIND-associated MD patterns, CT results and initiated VSP treatment. LP = L/P>40 occurs as an increase from baseline preceding a DIND, LP** = L/P>40 occurrs as an in-crease from a lower baseline during a complicated anesthesia and not associated with a DIND. D = a glutamine surge occurs during the DIND, A = a glutamine surge occurs after DIND, I = new ischemic lesion on CT, H = new hemorrhagic lesion on CT, 0 = no new lesion on CT. V = patient received VSP treatment.

Patie

nt #

Description of DIND

Isch

emic

MD

pa

ttern

glut

amin

e su

rge

at D

IND

CT

resu

lt

VSP

trea

tmen

t

1 RLS 3 to RLS 5 D 0 V 2 RLS 5 to RLS 8 A I 3 RLS 3 to RLS 7 LP A 0 V 4 Is RLS 5, increasing weakness left arm LP 0 5 RLS 5 to RLS 7 I 6 RLS 2 to RLS 6 LP A IH 7 RLS 3 to RLS 5 LP D 0 8 Is RLS 3, increasing weakness legs bilaterally D 0 V 9 RLS 3 to RLS 5 LP D IH 10 Is RLS 2, increasing agitation and disorientation LP D 0 V 11 Is RLS 1-2, onset of aphasia D 0 V 12 Is RLS 3, increasing right sided arm weakness A I 13 RLS 2 to RLS 3 and increasing right sided weakness D 0 V 14 RLS 3, increasing weakness right side and left leg LP** 0 17 RLS 3 to RLS 4, increasing left weakness, slow pupil reflexes D 0 V

Interstitial glutamine increase over time and glutamine surges may signal hard-working astrocytes In 28 out of 33 patients, glutamine levels were higher at the end of sampling compared with the beginning of the sampling and there was an overall sig-nificant increase in glutamine during the monitoring (Figure 14). Glutamine increases often came in sharp surges and this pattern was intriguing. In Paper IV and V one important aim was therefore to determine if there were clinical events associated with these glutamine surges.

45

Figure 14. The median interstitial glutamine concentration during the first 6 hours of sampling (start) and during the last 6 sampling hours (end) for each of the 33 pa-tients. The difference between each start and end value is also presented. Glutamine concentrations increased during the monitoring time (P<0.001).

DINDs are associated with glutamine surges Twelve out of 15 DINDs were associated with a glutamine surge (Table 1 and 2). In eight patients, the glutamine surge came during the deterioration and in four patients the glutamine surge came immediately following the DIND (Figure 15). Structural ischemia was absent on CT in seven DIND patients and these patients received VSP treatment. In five VSP-treated pa-tients the only DIND-associated MD pattern was a glutamine surge.

Neurons release glutamate at a lower ischemic threshold than astrocytes 66,120,124,166,176 and neuronal glutamate release accelerate astrocytic glutamate uptake, glycolysis and ATP utilization 107. We therefore suggest that the observed increase in glutamine and pyruvate during DIND is a favorable response in metabolically challenged tissue where the astrocytes maximize their glycolytic rate to yield ATP for glutamate uptake. Increasing glutamine and pyruvate levels in patients with suspected VSP, without ischemic MD-pattern and without ischemic CT signs, may consequently indicate increased neuronal glutamate release where the astrocytes increase the glutamate-glutamine cycle turnover by augmented glycolysis. As long as the astrocytes can accelerate the glutamate uptake the only clue that there is a high gluta-mate-glutamine cycle turnover will be increasing interstitial glutamine lev-els. The theory that astrocytes, by their relative higher resistance to oxygen and glucose deprivation, can rescue endangered neurons during ischemia, is supported by experimental data 201.

-500

0

500

1000

1500

2000

2500

glut

amin

e (�

M)

START DIFFERENCEEND

46

Figure 15. MD patterns in patient #8 who experienced a DIND during monitoring. Arrows indicate DIND onset. The DIND is associated with a glutamine surge and increasing pyruvate levels. The L/P ratio and glutamate are normal. See Table 1 and 2 for more details.

Glutamine surges appear when ICP is lowered Three out of our 33 patients had high mean hourly ICP-readings, in the 15-25 mmHg range, and their clinical deterioration or lack of improvement was partially attributed to the high ICP. The L/P ratio and glutamate were rela-tively normal during the period when ICP was high in these patients. To lower the ICP, CSF drainage at 10 mmHg was initiated. The lowering of ICP down to 10 mmHg was associated with sharply increasing glutamine and pyruvate levels in these three patients (Figure 16). We interpret this MD pattern as a sign of energy metabolic boost upon alleviation of a constraint that, in these three cases, was an unfavorably high ICP. One interpretation of these three cases is that moderately increased mean hourly ICP (15-25 mmHg) can hamper the brain tissue energy metabolism. This observation also generates a question whether SAH patients with moderately elevated ICP (15-20 mmHg) generally will benefit from CSF drainage at a lower pressure level than what is stated in current clinical treatment protocols.

20 40 60 80 100 120 140 160

hours after SAH

0

20

40

60

80

100

120

L/P

gl

utam

ate

(μM

)

0

50

100

150

200

250

300

350

pyru

vate

(μM

)

0

200

400

600

800

1000

glut

amin

e (μ

M)

47

Figure 16. Intracranial pressure (ICP), cerebral perfusion pressure (CPP), pyruvate and glutamine in one out of three patients where a glutamine surge was associated with ICP lowering achieved by increased CSF drainage. This patient underwent embolization of a left posterior cerebral artery aneurysm at monitoring hour 16 after which he was severely obtunded but obeyed command when pain stimulated. He had progressing right sided arm weakness from monitoring hour 38 and onwards. A CT-scan revealed suspected new ischemia in the territory of the left posterior cerebral artery. The ventricular drain was opened at 10 mmHg at monitoring hour 83 (arrow). As ICP decreased to 10 mmHg there was a surge in interstitial Gln and pyruvate.

Interstitial glutamine relate to the clinical admittance status Patients with poor WFNS grade at admittance (WFNS 4-5, n=16) had sig-nificantly lower glutamine at the onset of monitoring as compared to good WFNS grade patients (WFNS 1-3, n=17) (Figure 17). Glutamine levels 48 h after the SAH onset were higher in good WFNS grade patients (mean 486±174 μM, n=14) than in poor WFNS grade patients (mean 379±140μM, n=17) but this difference was not statistically significant (P=0.07). Mean glutamine levels during the monitoring was not significantly different (P=0.10) between the group with good recovery (WFNS 1-3 at the end of monitoring) as compared to those with worse recovery (WFNS 4-5 at the end of the monitoring time).

10 20 30 40 50 60 70 80 90 100 110 120 130 140

hours after SAH

4

6

8

10

12

14

16

ICP

(mm

Hg)

200

400

600

800

1000

1200

1400

1600

1800

glu

tam

ine

(μM

)

50

100

150

200

250

300

CPP

(mm

Hg)

py

ruva

te (μ

M)

48

Figure 17. Glutamine levels at onset of microdialysis monitoring in good WFNS grade (1-3) patients (n=17) and poor WFNS grade (4-5) patients (n=16). Glutamine levels at monitoring onset were lower in poor WFNS grade patients than in good WFNS patients (P=0.03). SE = standard error of the mean.

Cessation of sedation does not explain the glutamine surges Cessation of sedation and extubation were carried out in twelve out of the 33 patients during the monitoring. There were 23 glutamine surges in the entire material. Ten patients had both an extubation and a glutamine surge during the monitoring. In six of these ten cases the glutamine surge started within a time window of -12 h and +12 h from the time point of extubation. Overall, there was no association between presence of extubation during the monitor-ing and presence of a glutamine surge during the monitoring (P=0.46).

Intracranial hemodynamics and microdialysis measurements In Paper V the overall relationship between brain energy metabolism, the glutamate-glutamine cycle activity and intracranial hemodynamics was stud-ied. Correlation coefficients between MD variables and pressure parameters are presented in Table 3.

WFNS 1-3 (n=17) WFNS 4-5 (n=16)

Clincal grade at admission

300

400

500

600

mea

n gl

utam

ine

at m

onito

ring

onse

t (μM

)

Mean Mean ± SE Mean ± 1.96 SE

49

Table 3. Pearson linear correlation coefficients for hourly microdialysis measure-ments and pressure variables in 33 SAH patients receiving neurointensive care. Significant (P<0.05) correlation coefficients above 0.20 are presented.

MAP CPP ICP lactate pyruvate 0.23 0.24 L/P glutamate glutamine 0.25 -0.29 glutamine/glutamate -0.40

Glutamine and pyruvate correlated positively with CPP and these variables increased over time and during the NICU stay (Figure 14 and Figure 18). It is possible that an increasing interstitial concentration of pyruvate and glu-tamine, which was observed in the majority of the patients, is a consequence of improved cerebral perfusion. The relationship between CPP, MAP and pyruvate also provides additional indication, on a brain tissue metabolism level, that very high CPP may be beneficial following SAH 156. There was a negative correlation between glutamine and ICP (Figure 18C) and an even stronger correlation between glutamine/glutamate and ICP (Table 3). These observations are in line with the notion that low interstitial gluta-mine/glutamate ratios are associated with brain tissue constraint 145. MD patterns were generally better when ICP was below 10 mmHg as compared to hours when ICP exceeded 10 mmHg (Table 4). Previously it has been shown that SAH patients experiencing ICP above 20 mmHg have higher interstitial glutamate and L/P ratio levels as compared to patients without high ICP periods 161. High ICP (>20 mmHg) periods are also associated with clinical deterioration following SAH 156.

Table 4. Mean interstitial concentration during hours when ICP<10mmHg (n=780) and during hours when ICP>10mmHg (n=2723) in 33 SAH patients. For all metabo-lites the difference between the groups is statistically significant.

ICP<10 ICP>10 P L/P 25 29 <0.001glutamate μM 14 28 <0.001glutamine μM 748 565 <0.001glutamine/glutamate 382 132 <0.001lactate mM 3.6 4.1 <0.001pyruvate μM 148 142 <0.01

50

Figure 18. Scatterplots of hourly recordings in 33 SAH patients receiving neuroin-tensive care, (A) cerebral perfusion pressure vs. glutamine, (B) cerebral perfusion pressure vs. pyruvate, (C) intracranial pressure vs. glutamine.

-100 0 100 200 300 400 500

pyruvate (μM)

20

40

60

80

100

120

140

cere

bral

per

fusi

on p

ress

ure

(mm

Hg)

r = 0.24B

0 500 1000 1500 2000 2500

glutamine (μM)

-10

0

10

20

30

40

Intra

cran

ial p

ress

ure

(mm

Hg)

C r= - 0.29

-2000

200400

600800

10001200

14001600

18002000

22002400

26002800

glutamine (μM)

0

20

40

60

80

100

120

140

160

180

cere

bral

per

fusi

on p

ress

ure

(mm

Hg)

r = 0.25A

51

Interstitial glutamine � what does it mean? The glutamate-glutamine cycle is the major cerebral metabolic pathway for both glutamate and glutamine 67,181 89.

One important observation in this patient material was that fluctuations in glutamine levels over time were slower and less pronounced than for gluta-mate. This may not be surprising considering the high capacity systems for both glutamate release and glutamate uptake. Glutamine, on the other hand, travels more slowly, with both release into and uptake from the interstitial space partly depending on concentration gradients across the cell membranes of neurons and astrocytes 21. The absolute level of interstitial glutamine at any given time is therefore presumably a consequence of the astrocytic pro-duction rate, the rate of astrocytic release and the rate of neuronal uptake, during the preceding period.

Data on normal human brain interstitial glutamine levels is scarce. In pe-diatric TBI patients and in adult NICU patients where MD set-ups similar to ours were used 145 150, the highest glutamine level was 572 and 1200 μM, respectively. In animals, the reported basal brain interstitial glutamine con-centration range between 150 and 1000 μM 88,202. In view of this, glutamine levels above 1000 μM, which were present in a proportion of our observa-tions, are elevated. Ability to produce glutamine has been associated with neuronal survival when brain cells are subjected to stress 64,174 and injured brain tissue with high glutamate levels up-regulate GS 139. Such up-regulation may be one explanation behind the observed glutamine increase over time that occurred in the majority of the patients (Figure 14).

Reduced GS activity has been associated with increased interstitial gluta-mate levels despite normal amounts of EAAT-2 50. The very low glutamine levels early after SAH may be a sign of an overall low glutamate-glutamine cycling due to global ischemia associated with the bleed, with suppressed astrocyte energy metabolism and little glutamine synthesis due to lack of substrate and/or energy. Decreased glutamine uptake from blood could also be an explanation behind the low initial glutamine levels, since the energy requiring uptake from blood has been shown to decrease during ischemia 91.

Up to this point glutamine elevations in the patients have been explained by increased glutamate-glutamine cycle activity. However, alternative mechanisms behind the interstitial glutamine increases deserve considera-tion. One possible glutamine source is flow from the blood across the BBB. However, the glutamine transport from plasma to the brain interstitial space is normally highly regulated. It occurs in both directions but commonly with a net out-flux from the brain 73. In previous work and in work underway we have seen little correlation between alterations in plasma levels and brain interstitial levels for various amino acids, including glutamine and gluta-mate150. In addition, critical illness such as SAH is associated with decreas-ing plasma glutamine levels 175. This argues against that the observed fluc-

52

tuations in brain interstitial glutamine levels are significantly influenced by extracerebral sources.

Neurons depend on glutamine uptake for restoration of the presynaptic storage of vesicular transmitter glutamate 62,200. Decreased neuronal gluta-mine uptake would theoretically cause increased interstitial glutamine levels. However experimental data on SNAT 1-2 activity indicate that neuronal glutamine uptake is unaffected or even augmented in brain injury models 102,179.

When classifying the MD pattern a “glutamine surge” was defined as at least 150 μM increase in interstitial concentration over at least 12 h. Most often the increase during a surge was much greater than that. Twenty-three glutamine surges were present in the material. Retrospectively it was possi-ble to identify 15 clinical events, either a DIND (n=12) and/or a substantial decrease in ICP (n=3), that coincided with a glutamine surge. For the re-maining glutamine surges it was not possible to undoubtedly point out a corresponding event when examining the charts, write-ups and the digitally stored physiological data. However there is a range of possible influencing factors that were not taken into account in this study, e.g. minor alterations in CSF drainage levels and volumes, propofol and morphine require-ments/administration and improvements/deteriorations in consciousness level that were not picked up by the blunt RLS 85 grading.

One obvious hypothesis was that patients wakening up were the cause of the increasing glutamine levels. However neither was there an association between cessation of sedation/extubation and glutamine surges, nor was there a difference in glutamine levels at monitoring end (data not shown) or mean glutamine levels during the monitoring (Paper IV), between patients who woke up and those who did not.

Notably, the highest glutamine levels (>2 500 μM, Figure 11B, hour 133) in our material was not found in the metabolically most “normal” brain but in a brain with signs of intense energy production as well as a high intersti-tial glutamate load. This leaves us with a belief that the intriguing glutamine increase over time in a majority of the patients presumably represents hy-permetabolism in astrocytes that have been subjected to stress, including neuronal glutamate release.

Origin of low glutamine/glutamate ratios in animals and humans It is difficult to directly compare MD concentrations of metabolites obtained in animals and humans, mainly due to differences in probe membrane length, perfusion rate and the overall recovery. However, the glutamine/glutamate ratio and the L/P ratio are presumably recovery-independent 79.

53

The glutamate and glutamine levels measured in Paper II are in line with other works where 3 mm MD membrane probes and a perfusion rate of 2 μl/min were used 118,187. The glutamine/glutamate ratios in the iron-lesioned animals were always below 3.3 (see Paper II and Figure 8). In the human series only 3% of the monitored hours had glutamine/glutamate ratios below 3.4 and this was associated with overt ischemic L/P ratios (mean L/P ratio was 49 when glutamine/glutamate ratio was below 3.4). Iron-lesioned ani-mals with very low glutamine/glutamate ratios had mean L/P ratio below 30. This, together with the immunoblotting data, support that it is a structural impairment of GLT-1, and not energy failure, that is responsible for the ele-vated glutamate levels and the decreased glutamine levels at the iron-lesion.

Methodological and statistical considerations MD patterns are highly individualistic 122 and dependent on a range of fac-tors 79 why interpretation of real-time MD data is difficult at the bedside. In this thesis an attempt was made to determine how MD patterns relate to clinical events. An approach where MD data is visually observed, classified and correlated to clinical events, is difficult, time consuming and to some extent subjective. Nevertheless semi-quantitative observations of dynamic MD patterns and their relation to clinical course may be necessary in order to pursue beyond the existing knowledge that patients with overall abnormal MD measurements generally are worse off.

Many MD variables are not normally distributed and this appears to be especially true for glutamate where the brain has powerful mechanisms to maintain low interstitial levels. Once these mechanisms fail to work there are exponential increases in the interstitial concentration due to additive, patho-logical processes. Currently no consensus exists on how to treat MD data in statistical analyses and it is often at the discretion of the author. Throughout this thesis parametric and non-parametric statistics have been used on raw as well as on transformed data.

When exploring potential relationships in MD patterns, correlation analy-ses were done using all monitoring hours from all patients. There are prob-lems with such a practice since each of the 3600 monitoring hours was treated as an independent measure. This is not true, since a number of meas-urements are repeated and many measurements come from the same patient. It must also be considered that the hemodynamic parameters, to some extent, were controlled variables since intervention (e.g. CSF drainage) was initiated once they exceeded certain limits. These limitation and violations need to be considered when evaluating the presented results.

To summarize, this thesis is partly based on data from patients receiving intensive and individualized care. Observational and descriptive analyses were performed. This inherently involves a range of limitations. Neverthe-

54

less, in-vivo measurements from injured human brains have been presented together with careful analyses of clinical parameters. Hopefully these find-ings will lead to better questions, deeper excursions and ultimately improved understanding of pathophysiological and recovering processes in the injured brain. Furthermore, the combination of MD and advanced neuroimaging (e.g. magnetic resonance spectroscopy and positron emission tomography), not readily available in this study may further potentiate the process of im-proved understanding of the neurochemistry and pathophysiology of acute brain injury.

55

Concluding remarks

This thesis spans over animal experiments and clinical observations on glu-tamate turnover in the injured brain. The major findings are:

Astrocytic glutamate uptake protein GLT-1 is structurally impaired and ag-gregated in iron-induced epileptogenic cortical lesions in rat, probably as a consequence of oxidative damage.

Reduced monomeric GLT-1 levels are associated with increased interstitial glutamate levels and decreased interstitial glutamine levels at the iron-lesion in rat. These amino acid alterations are most likely a consequence of de-creased astrocytic glutamate uptake and may play a role in the development of posttraumatic epilepsy.

In human SAH, the brain interstitial glutamine/glutamate ratio correlates inversely with the L/P ratio. Glutamine increase over time and there is a correlation between glutamine and pyruvate. During periods of energy cri-ses, defined as L/P ratio above 40, glutamine will increase or decrease de-pending on whether pyruvate is high or low.

Low interstitial pyruvate levels may signal a reduced energy producing ca-pacity, unrelated to ischemia, which hampers the energy demanding gluta-mate-glutamine cycling.

Interstitial L/P ratios above 40 following SAH is a pattern associated with functional and/or structural ischemia. Delayed ischemic neurological deficits are associated with L/P ratio above 40 and/or surges in interstitial glutamine. Surges in interstitial glutamine predominately occur during delayed ischemic deteriorations where high L/P ratios and radiological ischemia are absent.

Cerebral perfusion pressure correlates with interstitial pyruvate and gluta-mine. Intracranial pressure correlates with glutamine and the gluta-mine/glutamate ratio. Glutamine surges appeared when a high ICP was sub-stantially lowered by increased CSF drainage.

56

Acknowledgements

Thank you... Elisabeth Ronne-Engström for enthusiasm, support, determination and friendship. I have learned so much and made many valuable experiences during this project. I admire you for many things. Thank you. Lars Hillered for being a bottom-less source of knowledge, ideas and scien-tific inspiration. Eva Kumlien for important discussions, perspectives, advice and encour-agement when needed. Dan Lindholm for letting me work in your laboratory. Tim Howells, father of Browser and co-author, for fruitful discussions on data handling and for filling the workplace with pleasant presence and mu-sic. Per Enblad and Pelle Nilsson for valuable co-authoring and a strong interest in neurointensive care research. Co-authors; Åse Elfving, Roland Flink, Göran Hesselager, Anders Lewén, Niklas Marklund, Mats Ryttlefors, Konstantin Salci and Maria Zetterling. Inger Ståhl Myllyaho, without whose careful, expertise work this would only be half a thesis. Lennart Persson and Hans Bolander for research support. Urban Ungerstedt for allowing your facilities to some of my analyses. Maude Karlsdotter for assistance and supreme administration of our unit. Kristin Elf, indispensable former PhD-colleague, who I appreciate im-mensely.

57

Fredrik Clausen for great help during the animal-phase. All personnel at NIVA and Neuro-Op/An for good care of the patients. Torbjörn Karlsson, Lars Wiklund and Torsten Gordh, for appreciation of my research. For provision of research resources and time. Anestesi- och intensivvårdskliniken, Halmstad Länssjukhus, for research time and for a positive attitude towards me doing research.

Clinical tutors and role-models, in particular; Birgitta Birgisdottir, Susan Erichsen, Rainer Dörenberg, Jakob Johansson, Miklós Lipcsey, Emma Pon-ten and Johann Valtysson.

Lennart Christiansson, outstanding Residency Program Director, for answers and for inspiration to become excellent. Swedish social welfare in general and Förskolan Tittut in particular. Friends – I am blessed with a large number of the best kind and I love to be around you. Thank you for visits, invitations, meals, chats, walks, excur-sions, conversations, inspiration, advice and support. Ann-Kristin and Kjell, probably the world’s best parents-in-law. My sister Josefine and brother Anthon. So different – but equally terrific! My parents Ingrid and Bernt for massive, invaluable help and support in various kinds and shapes. Cesar, Svea and Jussi, the most important and precious I ever had and ever did. Daniel, for reading me like a book and playing me like an instrument. You are loved.

... curiosity and perseverance.

58

References

1. Adams JH, Graham DI, Jennett B: The structural basis of moderate disabil-ity after traumatic brain damage. J Neurol Neurosurg Psychiatry 71:521-524, 2001

2. Akbar MT, Rattray M, Williams RJ, Chong NW, Meldrum BS: Reduction of GABA and glutamate transporter messenger RNAs in the severe-seizure genetically epilepsy-prone rat. Neuroscience 85:1235-1251, 1998

3. Akbar MT, Torp R, Danbolt NC, Levy LM, Meldrum BS, Ottersen OP: Expression of glial glutamate transporters GLT-1 and GLAST is unchanged in the hippocampus in fully kindled rats. Neuroscience 78:351-359, 1997

4. Amara SG, Fontana AC: Excitatory amino acid transporters: keeping up with glutamate. Neurochem Int 41:313-318, 2002

5. Anderson JJ, DiMicco JA: The use of microdialysis for studying the re-gional effects of pharmacological manipulation on extracellular levels of amino acids--some methodological aspects. Life Sci 51:623-630, 1992

6. Andersson EH, Björklund R, Emanuelson I, Stålhammar D: Epidemiology of traumatic brain injury: a population based study in western Sweden. Acta Neurol Scand 107:256-259, 2003

7. Angeleri F, Majkowski J, Cacchio G, Sobieszek A, D'Acunto S, Gesuita R, et al: Posttraumatic epilepsy risk factors: one-year prospective study after head injury. Epilepsia 40:1222-1230, 1999

8. Annegers JF, Coan SP: The risks of epilepsy after traumatic brain injury. Seizure 9:453-457, 2000

9. Arboix A, Garcia-Eroles L, Massons JB, Oliveres M, Comes E: Predictive factors of early seizures after acute cerebrovascular disease. Stroke 28:1590-1594, 1997

10. Arita K, Kurisu K, Iida K, Hanaya R, Sugiyama K, Akimitsu T, et al: Sur-gical treatment for intractable epilepsy caused by cavernous angioma in the temporal lobe of the dominant hemisphere--three case reports. Neurol Med Chir (Tokyo) 40:439-445, 2000

11. Asikainen I, Kaste M, Sarna S: Early and late posttraumatic seizures in traumatic brain injury rehabilitation patients: brain injury factors causing late seizures and influence of seizures on long-term outcome. Epilepsia 40:584-589, 1999

12. Bartnik BL, Lee SM, Hovda DA, Sutton RL: The fate of glucose during the period of decreased metabolism after fluid percussion injury: a 13C NMR study. J Neurotrauma 24:1079-1092, 2007

13. Bellander BM, Cantais E, Enblad P, Hutchinson P, Nordström CH, Robert-son C, et al: Consensus meeting on microdialysis in neurointensive care. Intensive Care Med 30:2166-2169, 2004

59

14. Bendok BR, Getch CC, Malisch TW, Batjer HH: Treatment of aneurysmal subarachnoid hemorrhage. Semin Neurol 18:521-531, 1998

15. Benveniste H: Brain microdialysis. J Neurochem 52:1667-1679, 1989 16. Benveniste H, Hansen AJ, Ottosen NS: Determination of brain interstitial

concentrations by microdialysis. J Neurochem 52:1741-1750, 1989 17. Bergersen LH: Is lactate food for neurons? Comparison of monocarboxy-

late transporter subtypes in brain and muscle. Neuroscience 145:11-19, 2007

18. Boulland JL, Osen KK, Levy LM, Danbolt NC, Edwards RH, Storm-Mathisen J, et al: Cell-specific expression of the glutamine transporter SN1 suggests differences in dependence on the glutamine cycle. Eur J Neurosci 15:1615-1631, 2002

19. Bradford HF: Glutamate, GABA and epilepsy. Prog Neurobiol 47:477-511, 1995

20. Broer A, Albers A, Setiawan I, Edwards RH, Chaudhry FA, Lang F, et al: Regulation of the glutamine transporter SN1 by extracellular pH and intra-cellular sodium ions. J Physiol 539:3-14, 2002

21. Broer S, Brookes N: Transfer of glutamine between astrocytes and neurons. J Neurochem 77:705-719, 2001

22. Brookes N: Effect of intracellular glutamine on the uptake of large neutral amino acids in astrocytes: concentrative Na(+)-independent transport ex-hibits metastability. J Neurochem 59:227-235, 1992

23. Bruns J, Jr., Hauser WA: The epidemiology of traumatic brain injury: a review. Epilepsia 44 Suppl 10:2-10, 2003

24. Buck CR, Jurynec MJ, Gupta DK, Law AK, Bilger J, Wallace DC, et al: Increased adenine nucleotide translocator 1 in reactive astrocytes facilitates glutamate transport. Exp Neurol 181:149-158, 2003

25. Bullock R, Zauner A, Woodward JJ, Myseros J, Choi SC, Ward JD, et al: Factors affecting excitatory amino acid release following severe human head injury. J Neurosurg 89:507-518, 1998

26. Butzkueven H, Evans AH, Pitman A, Leopold C, Jolley DJ, Kaye AH, et al: Onset seizures independently predict poor outcome after subarachnoid hemorrhage. Neurology 55:1315-1320, 2000

27. Byard RW, Schliebs J, Koszyca BA: Osler-Weber-Rendu syndrome--pathological manifestations and autopsy considerations. J Forensic Sci 46:698-701, 2001

28. Cao X, Phillis JW: alpha-Phenyl-tert-butyl-nitrone reduces cortical infarct and edema in rats subjected to focal ischemia. Brain Res 644:267-272, 1994

29. Cerdan S, Rodrigues TB, Sierra A, Benito M, Fonseca LL, Fonseca CP, et al: The redox switch/redox coupling hypothesis. Neurochem Int 48:523-530, 2006

30. Chadwick DW: Seizures and epilepsy in adults, in Laidlaw J, Richens A, Chadwick DW (eds): A textbook of epilepsy, ed 4th. Edinburgh: Churchill Livingstone, 1993, pp 177-183

31. Chapman AG: Glutamate and epilepsy. J Nutr 130:1043S-1045S, 2000 32. Chen G, Griffin M, Poyer JL, McCay PB: HPLC procedure for the pharma-

cokinetic study of the spin-trapping agent, alpha-phenyl-N-tert-butyl ni-trone (PBN). Free Radic Biol Med 9:93-98, 1990

60

33. Cheng HY, Liu T, Feuerstein G, Barone FC: Distribution of spin-trapping compounds in rat blood and brain: in vivo microdialysis determination. Free Radic Biol Med 14:243-250, 1993

34. Chesnut RM: Care of central nervous system injuries. Surg Clin North Am 87:119-156, vii, 2007

35. Chu K, Lee ST, Sinn DI, Ko SY, Kim EH, Kim JM, et al: Pharmacological Induction of Ischemic Tolerance by Glutamate Transporter-1 (EAAT2) Upregulation. Stroke 38:177-182, 2007

36. Claassen J, Peery S, Kreiter KT, Hirsch LJ, Du EY, Connolly ES, et al: Predictors and clinical impact of epilepsy after subarachnoid hemorrhage. Neurology 60:208-214, 2003

37. Clements JD, Westbrook GL: Activation kinetics reveal the number of glutamate and glycine binding sites on the N-methyl-D-aspartate receptor. Neuron 7:605-613, 1991

38. Conti F, Melone M: The glutamine commute: lost in the tube? Neurochem Int 48:459-464, 2006

39. Dabrosin C, Hallstrom A, Ungerstedt U, Hammar M: Microdialysis of human breast tissue during the menstrual cycle. Clin Sci (Lond) 92:493-496, 1997

40. Daikhin Y, Yudkoff M: Compartmentation of brain glutamate metabolism in neurons and glia. J Nutr 130:1026S-1031S, 2000

41. Danbolt NC: Glutamate uptake. Prog Neurobiol 65:1-105, 2001 42. Danbolt NC, Storm-Mathisen J, Kanner BI: An [Na+ + K+]coupled L-

glutamate transporter purified from rat brain is located in glial cell proc-esses. Neuroscience 51:295-310, 1992

43. De Micheli E, Pinna G, Alfieri A, Caramia G, Bianchi L, Colivicchi MA, et al: Post-operative monitoring of cortical taurine in patients with subarachnoid hemorrhage: a microdialysis study. Adv Exp Med Biol 483:595-603, 2000

44. Dennis LJ, Claassen J, Hirsch LJ, Emerson RG, Connolly ES, Mayer SA: Nonconvulsive status epilepticus after subarachnoid hemorrhage. Neuro-surgery 51:1136-1143; discussion 1144, 2002

45. DeWitt DS, Jenkins LW, Prough DS: Enhanced vulnerability to secondary ischemic insults after experimental traumatic brain injury. New Horiz 3:376-383, 1995

46. Dolinska M, Zablocka B, Sonnewald U, Albrecht J: Glutamine uptake and expression of mRNA's of glutamine transporting proteins in mouse cerebel-lar and cerebral cortical astrocytes and neurons. Neurochem Int 44:75-81, 2004

47. Dorhout Mees SM, Rinkel GJ, Feigin VL, Algra A, van den Bergh WM, Vermeulen M, et al: Calcium antagonists for aneurysmal subarachnoid haemorrhage. Cochrane Database Syst Rev:CD000277, 2007

48. During MJ, Spencer DD: Extracellular hippocampal glutamate and sponta-neous seizure in the conscious human brain. Lancet 341:1607-1610, 1993

49. Dusick JR, Glenn TC, Lee WN, Vespa PM, Kelly DF, Lee SM, et al: In-creased pentose phosphate pathway flux after clinical traumatic brain in-jury: a [1,2-13C2]glucose labeling study in humans. J Cereb Blood Flow Metab 27:1593-1602, 2007

61

50. Eid T, Thomas MJ, Spencer DD, Runden-Pran E, Lai JC, Malthankar GV, et al: Loss of glutamine synthetase in the human epileptogenic hippocam-pus: possible mechanism for raised extracellular glutamate in mesial tem-poral lobe epilepsy. Lancet 363:28-37, 2004

51. Enblad P, Frykholm P, Valtysson J, Silander HC, Andersson J, Fasth KJ, et al: Middle cerebral artery occlusion and reperfusion in primates monitored by microdialysis and sequential positron emission tomography. Stroke 32:1574-1580, 2001

52. Enblad P, Persson L: Impact on clinical outcome of secondary brain insults during the neurointensive care of patients with subarachnoid haemorrhage: a pilot study. J Neurol Neurosurg Psychiatry 62:512-516, 1997

53. Enblad P, Valtysson J, Andersson J, Lilja A, Valind S, Antoni G, et al: Simultaneous intracerebral microdialysis and positron emission tomogra-phy in the detection of ischemia in patients with subarachnoid hemorrhage. J Cereb Blood Flow Metab 16:637-644, 1996

54. Erecinska M, Silver IA: Metabolism and role of glutamate in mammalian brain. Prog Neurobiol 35:245-296, 1990

55. Eriksson G, Peterson A, Iverfeldt K, Walum E: Sodium-dependent gluta-mate uptake as an activator of oxidative metabolism in primary astrocyte cultures from newborn rat. Glia 15:152-156, 1995

56. Faden AI, Demediuk P, Panter SS, Vink R: The role of excitatory amino acids and NMDA receptors in traumatic brain injury. Science 244:798-800, 1989

57. Farrell MA, Vinters HV: General neuropathology of epilepsy, in Engel J, Jr., Pedley TA (eds): Epilepsy: a comprehensive textbook. Philadelphia: Lippincott-Raven Publishers, 1997, Vol 1, pp 157-175

58. Faul M, Wald MM, Rutland-Brown W, Sullivent EE, Sattin RW: Using a cost-benefit analysis to estimate outcomes of a clinical treatment guideline: testing theBrain Trauma Foundation guidelines for the treatment of severe traumatic brain injury. J Trauma 63:1271-1278, 2007

59. Fedele E, Mazzone P, Stefani A, Bassi A, Ansaldo MA, Raiteri M, et al: Microdialysis in Parkinsonian patient basal ganglia: acute apomorphine-induced clinical and electrophysiological effects not paralleled by changes in the release of neuroactive amino acids. Exp Neurol 167:356-365, 2001

60. Frey LC: Epidemiology of posttraumatic epilepsy: a critical review. Epi-lepsia 44 Suppl 10:11-17, 2003

61. Gemba T, Oshima T, Ninomiya M: Glutamate efflux via the reversal of the sodium-dependent glutamate transporter caused by glycolytic inhibition in rat cultured astrocytes. Neuroscience 63:789-795, 1994

62. Gibbs ME, O'Dowd BS, Hertz L, Robinson SR, Sedman GL, Ng KT: Inhi-bition of glutamine synthetase activity prevents memory consolidation. Brain Res Cogn Brain Res 4:57-64, 1996

63. Golding EM, Robertson CS, Bryan RM, Jr.: The consequences of traumatic brain injury on cerebral blood flow and autoregulation: a review. Clin Exp Hypertens 21:299-332, 1999

64. Gorovits R, Avidan N, Avisar N, Shaked I, Vardimon L: Glutamine syn-thetase protects against neuronal degeneration in injured retinal tissue. Proc Natl Acad Sci U S A 94:7024-7029, 1997

62

65. Haltiner AM, Temkin NR, Dikmen SS: Risk of seizure recurrence after the first late posttraumatic seizure. Arch Phys Med Rehabil 78:835-840, 1997

66. Hamann M, Rossi DJ, Marie H, Attwell D: Knocking out the glial gluta-mate transporter GLT-1 reduces glutamate uptake but does not affect hip-pocampal glutamate dynamics in early simulated ischaemia. Eur J Neuro-sci 15:308-314, 2002

67. Hamberger AC, Chiang GH, Nylen ES, Scheff SW, Cotman CW: Gluta-mate as a CNS transmitter. I. Evaluation of glucose and glutamine as pre-cursors for the synthesis of preferentially released glutamate. Brain Res 168:513-530, 1979

68. Harper AM: Autoregulation of cerebral blood flow: influence of the arterial blood pressure on the blood flow through the cerebral cortex. J Neurol Neurosurg Psychiatry 29:398-403, 1966

69. Harrod CG, Bendok BR, Batjer HH: Prediction of cerebral vasospasm in patients presenting with aneurysmal subarachnoid hemorrhage: a review. Neurosurgery 56:633-654; discussion 633-654, 2005

70. Hashizume K, Takizawa K, Kunimoto M, Yoshida K, Tanaka T, Yonemasu Y, et al: [Hippocampal hemosiderin deposit due to large pituitary adenoma presenting temporal lobe epilepsy--a case report]. No To Shinkei 49:366-371, 1997

71. Haugeto O, Ullensvang K, Levy LM, Chaudhry FA, Honore T, Nielsen M, et al: Brain glutamate transporter proteins form homomultimers. J Biol Chem 271:27715-27722, 1996

72. Hauser WA: Incidence and prevalence, in Engel JJ, Pedley TA (eds): Epi-lepsy: A comprehensive textbook, ed 4th. Philadelphia: Lippincott-Raven, 1997, pp 47-57

73. Hawkins RA, O'Kane RL, Simpson IA, Vina JR: Structure of the blood-brain barrier and its role in the transport of amino acids. J Nutr 136:218S-226S, 2006

74. Heikkinen ER, Ronty HS, Tolonen U, Pyhtinen J: Development of post-traumatic epilepsy. Stereotact Funct Neurosurg 54-55:25-33, 1990

75. Herman MA, Jahr CE: Extracellular glutamate concentration in hippocam-pal slice. J Neurosci 27:9736-9741, 2007

76. Hertz L: Glutamate, a neurotransmitter--and so much more. A synopsis of Wierzba III. Neurochem Int 48:416-425, 2006

77. Hillered L, Persson L, Nilsson P, Ronne-Engström E, Enblad P: Continu-ous monitoring of cerebral metabolism in traumatic brain injury: a focus on cerebral microdialysis. Curr Opin Crit Care 12:112-118, 2006

78. Hillered L, Persson L, Ponten U, Ungerstedt U: Neurometabolic monitoring of the ischaemic human brain using microdialysis. Acta Neurochir (Wien) 102:91-97, 1990

79. Hillered L, Vespa PM, Hovda DA: Translational neurochemical research in acute human brain injury: the current status and potential future for cerebral microdialysis. J Neurotrauma 22:3-41, 2005

80. Hisada K, Morioka T, Nishio S, Amano T, Iwaki T, Fukui M: [Temporal lobe epilepsy associated with old intracerebral hemorrhage due to capillary telangiectasis in the temporal lobe: case report]. No To Shinkei 51:729-735, 1999

63

81. Hlatky R, Valadka AB, Goodman JC, Contant CF, Robertson CS: Patterns of energy substrates during ischemia measured in the brain by microdialy-sis. J Neurotrauma 21:894-906, 2004

82. Howells T, Elf K, Jones PA, Ronne-Engström E, Piper I, Nilsson P, et al: Pressure reactivity as a guide in the treatment of cerebral perfusion pressure in patients with brain trauma. J Neurosurg 102:311-317, 2005

83. Hutchinson PJ, Gupta AK, Fryer TF, Al-Rawi PG, Chatfield DA, Coles JP, et al: Correlation between cerebral blood flow, substrate delivery, and me-tabolism in head injury: a combined microdialysis and triple oxygen posi-tron emission tomography study. J Cereb Blood Flow Metab 22:735-745, 2002

84. Hutchinson PJ, O'Connell MT, Al-Rawi PG, Maskell LB, Kett-White R, Gupta AK, et al: Clinical cerebral microdialysis: a methodological study. J Neurosurg 93:37-43, 2000

85. Ikematsu K, Tsuda R, Kondo T, Nakasono I: The expression of excitatory amino acid transporter 2 in traumatic brain injury. Forensic Sci Int 130:83-89, 2002

86. Kabuto H, Yokoi I, Ogawa N: Melatonin inhibits iron-induced epileptic discharges in rats by suppressing peroxidation. Epilepsia 39:237-243, 1998

87. Kanamori K, Ross BD: Kinetics of glial glutamine efflux and the mecha-nism of neuronal uptake studied in vivo in mildly hyperammonemic rat brain. J Neurochem 99:1103-1113, 2006

88. Kanamori K, Ross BD: Quantitative determination of extracellular gluta-mine concentration in rat brain, and its elevation in vivo by system A trans-port inhibitor, alpha-(methylamino)isobutyrate. J Neurochem 90:203-210, 2004

89. Kanamori K, Ross BD, Kondrat RW: Glial uptake of neurotransmitter glutamate from the extracellular fluid studied in vivo by microdialysis and (13)C NMR. J Neurochem 83:682-695, 2002

90. Kasischke KA, Vishwasrao HD, Fisher PJ, Zipfel WR, Webb WW: Neural activity triggers neuronal oxidative metabolism followed by astrocytic gly-colysis. Science 305:99-103, 2004

91. Kawai N, Stummer W, Ennis SR, Betz AL, Keep RF: Blood-brain barrier glutamine transport during normoglycemic and hyperglycemic focal cere-bral ischemia. J Cereb Blood Flow Metab 19:79-86, 1999

92. Kehr J: A survey on quantitative microdialysis: theoretical models and practical implications. J Neurosci Methods 48:251-261, 1993

93. Keller JN, Mark RJ, Bruce AJ, Blanc E, Rothstein JD, Uchida K, et al: 4-Hydroxynonenal, an aldehydic product of membrane lipid peroxidation, impairs glutamate transport and mitochondrial function in synaptosomes. Neuroscience 80:685-696, 1997

94. Kett-White R, Hutchinson PJ, al-Rawi PG, Gupta AK, O'Connell MT, Pickard JD, et al: Extracellular lactate/pyruvate and glutamate changes in patients during per-operative episodes of cerebral ischaemia. Acta Neuro-chir Suppl 81:363-365, 2002

95. Komatsu M, Hiramatsu M: The efficacy of an antioxidant cocktail on lipid peroxide level and superoxide dismutase activity in aged rat brain and DNA damage in iron-induced epileptogenic foci. Toxicology 148:143-148, 2000

64

96. Kosugi T, Kawahara K: Reversed actrocytic GLT-1 during ischemia is crucial to excitotoxic death of neurons, but contributes to the survival of as-trocytes themselves. Neurochem Res 31:933-943, 2006

97. Kotila M, Waltimo O: Epilepsy after stroke. Epilepsia 33:495-498, 1992 98. Kumar R, Gupta RK, Husain M, Vatsal DK, Chawla S, Rathore RK, et al:

Magnetization transfer MR imaging in patients with posttraumatic epilepsy. AJNR Am J Neuroradiol 24:218-224, 2003

99. Lam JM, Smielewski P, Czosnyka M, Pickard JD, Kirkpatrick PJ: Predict-ing delayed ischemic deficits after aneurysmal subarachnoid hemorrhage using a transient hyperemic response test of cerebral autoregulation. Neu-rosurgery 47:819-825; discussions 825-816, 2000

100. Lanzino G, Kassell NF: Double-blind, randomized, vehicle-controlled study of high-dose tirilazad mesylate in women with aneurysmal subarach-noid hemorrhage. Part II. A cooperative study in North America. J Neuro-surg 90:1018-1024, 1999

101. Lanzino G, Kassell NF, Dorsch NW, Pasqualin A, Brandt L, Schmiedek P, et al: Double-blind, randomized, vehicle-controlled study of high-dose tiri-lazad mesylate in women with aneurysmal subarachnoid hemorrhage. Part I. A cooperative study in Europe, Australia, New Zealand, and South Af-rica. J Neurosurg 90:1011-1017, 1999

102. Leibovici A, Rossignol C, Montrowl JA, Erickson JD, Varoqui H, Wata-nabe M, et al: The effects of hypoxia-ischemia on neutral amino acid trans-porters in the developing rat brain. Dev Neurosci 29:268-274, 2007

103. Lerma J, Herranz AS, Herreras O, Abraira V, Martin del Rio R: In vivo determination of extracellular concentration of amino acids in the rat hip-pocampus. A method based on brain dialysis and computerized analysis. Brain Res 384:145-155, 1986

104. Lin AM, Chyi BY, Wang SD, Yu HH, Kanakamma PP, Luh TY, et al: Carboxyfullerene prevents iron-induced oxidative stress in rat brain. J Neurochem 72:1634-1640, 1999

105. Lindefors N, Amberg G, Ungerstedt U: Intracerebral microdialysis: I. Ex-perimental studies of diffusion kinetics. J Pharmacol Methods 22:141-156, 1989

106. Lönnroth P, Jansson PA, Smith U: A microdialysis method allowing char-acterization of intercellular water space in humans. Am J Physiol 253:E228-231, 1987

107. Magistretti PJ, Pellerin L: Astrocytes Couple Synaptic Activity to Glucose Utilization in the Brain. News Physiol Sci 14:177-182, 1999

108. Magistretti PJ, Pellerin L: Cellular mechanisms of brain energy metabolism and their relevance to functional brain imaging. Philos Trans R Soc Lond B Biol Sci 354:1155-1163, 1999

109. Magistretti PJ, Pellerin L, Rothman DL, Shulman RG: Energy on demand. Science 283:496-497, 1999

110. Mallolas J, Hurtado O, Castellanos M, Blanco M, Sobrino T, Serena J, et al: A polymorphism in the EAAT2 promoter is associated with higher glu-tamate concentrations and higher frequency of progressing stroke. J Exp Med 203:711-717, 2006

111. Marklund N, Lewander T, Clausen F, Hillered L: Effects of the nitrone radical scavengers PBN and S-PBN on in vivo trapping of reactive oxygen

65

species after traumatic brain injury in rats. J Cereb Blood Flow Metab 21:1259-1267, 2001

112. Masson F, Thicoipe M, Aye P, Mokni T, Senjean P, Schmitt V, et al: Epi-demiology of severe brain injuries: a prospective population-based study. J Trauma 51:481-489, 2001

113. McKenna MC, Sonnewald U, Huang X, Stevenson J, Zielke HR: Exoge-nous glutamate concentration regulates the metabolic fate of glutamate in astrocytes. J Neurochem 66:386-393, 1996

114. Meldrum BS: The role of glutamate in epilepsy and other CNS disorders. Neurology 44:S14-23, 1994

115. Meldrum BS, Akbar MT, Chapman AG: Glutamate receptors and trans-porters in genetic and acquired models of epilepsy. Epilepsy Res 36:189-204, 1999

116. Menacherry S, Hubert W, Justice JB, Jr.: In vivo calibration of microdialy-sis probes for exogenous compounds. Anal Chem 64:577-583, 1992

117. Meyerson BA, Linderoth B, Karlsson H, Ungerstedt U: Microdialysis in the human brain: extracellular measurements in the thalamus of parkinsonian patients. Life Sci 46:301-308, 1990

118. Michalak A, Rose C, Butterworth J, Butterworth RF: Neuroactive amino acids and glutamate (NMDA) receptors in frontal cortex of rats with ex-perimental acute liver failure. Hepatology 24:908-913, 1996

119. Miele M, Berners M, Boutelle MG, Kusakabe H, Fillenz M: The determi-nation of the extracellular concentration of brain glutamate using quantita-tive microdialysis. Brain Res 707:131-133, 1996

120. Mitani A, Tanaka K: Functional changes of glial glutamate transporter GLT-1 during ischemia: an in vivo study in the hippocampal CA1 of nor-mal mice and mutant mice lacking GLT-1. J Neurosci 23:7176-7182, 2003

121. Moriwaki A, Hattori Y, Hayashi Y, Hori Y: Development of epileptic ac-tivity induced by iron injection into rat cerebral cortex: electrographic and behavioral characteristics. Electroencephalogr Clin Neurophysiol 83:281-288, 1992

122. Nelson DW, Bellander BM, Maccallum RM, Axelsson J, Alm M, Wallin M, et al: Cerebral microdialysis of patients with severe traumatic brain in-jury exhibits highly individualistic patterns as visualized by cluster analysis with self-organizing maps. Crit Care Med 32:2428-2436, 2004

123. Nilsson P, Hillered L, Ponten U, Ungerstedt U: Changes in cortical ex-tracellular levels of energy-related metabolites and amino acids following concussive brain injury in rats. J Cereb Blood Flow Metab 10:631-637, 1990

124. Nishizawa Y: Glutamate release and neuronal damage in ischemia. Life Sci 69:369-381, 2001

125. Ogden JA, Utley T, Mee EW: Neurological and psychosocial outcome 4 to 7 years after subarachnoid hemorrhage. Neurosurgery 41:25-34, 1997

126. Olney JW, Sharpe LG: Brain lesions in an infant rhesus monkey treated with monsodium glutamate. Science 166:386-388, 1969

127. Patel AJ, Hunt A: Concentration of free amino acids in primary cultures of neurones and astrocytes. J Neurochem 44:1816-1821, 1985

66

128. Patneau DK, Mayer ML: Structure-activity relationships for amino acid transmitter candidates acting at N-methyl-D-aspartate and quisqualate re-ceptors. J Neurosci 10:2385-2399, 1990

129. Paulson OB, Strandgaard S, Edvinsson L: Cerebral autoregulation. Cere-brovasc Brain Metab Rev 2:161-192, 1990

130. Pawlak J, Brito V, Kuppers E, Beyer C: Regulation of glutamate trans-porter GLAST and GLT-1 expression in astrocytes by estrogen. Brain Res Mol Brain Res 138:1-7, 2005

131. Peeling J, Yan HJ, Chen SG, Campbell M, Del Bigio MR: Protective ef-fects of free radical inhibitors in intracerebral hemorrhage in rat. Brain Res 795:63-70, 1998

132. Pellerin L, Bouzier-Sore AK, Aubert A, Serres S, Merle M, Costalat R, et al: Activity-dependent regulation of energy metabolism by astrocytes: An update. Glia 55:1251-1262, 2007

133. Peng L, Swanson RA, Hertz L: Effects of L-glutamate, D-aspartate, and monensin on glycolytic and oxidative glucose metabolism in mouse astro-cyte cultures: further evidence that glutamate uptake is metabolically driven by oxidative metabolism. Neurochem Int 38:437-443, 2001

134. Persson L, Hillered L: Chemical monitoring of neurosurgical intensive care patients using intracerebral microdialysis. J Neurosurg 76:72-80, 1992

135. Persson L, Valtysson J, Enblad P, Wärme PE, Cesarini K, Lewen A, et al: Neurochemical monitoring using intracerebral microdialysis in patients with subarachnoid hemorrhage. J Neurosurg 84:606-616, 1996

136. Petroff OA, Errante LD, Rothman DL, Kim JH, Spencer DD: Glutamate-glutamine cycling in the epileptic human hippocampus. Epilepsia 43:703-710, 2002

137. Phillis JW, Ren J, O'Regan MH: Transporter reversal as a mechanism of glutamate release from the ischemic rat cerebral cortex: studies with DL-threo-beta-benzyloxyaspartate. Brain Res 868:105-112, 2000

138. Pyne-Geithman GJ, Morgan CJ, Wagner K, Dulaney EM, Carrozzella J, Kanter DS, et al: Bilirubin production and oxidation in CSF of patients with cerebral vasospasm after subarachnoid hemorrhage. J Cereb Blood Flow Metab 25:1070-1077, 2005

139. Ramonet D, Rodriguez MJ, Fredriksson K, Bernal F, Mahy N: In vivo neuroprotective adaptation of the glutamate/glutamine cycle to neuronal death. Hippocampus 14:586-594, 2004

140. Rao VL, Baskaya MK, Dogan A, Rothstein JD, Dempsey RJ: Traumatic brain injury down-regulates glial glutamate transporter (GLT-1 and GLAST) proteins in rat brain. J Neurochem 70:2020-2027, 1998

141. Rao VL, Dogan A, Bowen KK, Todd KG, Dempsey RJ: Antisense knock-down of the glial glutamate transporter GLT-1 exacerbates hippocampal neuronal damage following traumatic injury to rat brain. Eur J Neurosci 13:119-128, 2001

142. Rao VL, Dogan A, Todd KG, Bowen KK, Kim BT, Rothstein JD, et al: Antisense knockdown of the glial glutamate transporter GLT-1, but not the neuronal glutamate transporter EAAC1, exacerbates transient focal cerebral ischemia-induced neuronal damage in rat brain. J Neurosci 21:1876-1883, 2001

67

143. Reinstrup P, Ståhl N, Mellergard P, Uski T, Ungerstedt U, Nordström CH: Intracerebral microdialysis in clinical practice: baseline values for chemical markers during wakefulness, anesthesia, and neurosurgery. Neurosurgery 47:701-709; discussion 709-710, 2000

144. Rhoney DH, Tipps LB, Murry KR, Basham MC, Michael DB, Coplin WM: Anticonvulsant prophylaxis and timing of seizures after aneurysmal su-barachnoid hemorrhage. Neurology 55:258-265, 2000

145. Richards DA, Tolias CM, Sgouros S, Bowery NG: Extracellular glutamine to glutamate ratio may predict outcome in the injured brain: a clinical mi-crodialysis study in children. Pharmacol Res 48:101-109, 2003

146. Riksstroke Sf: RIKS-STROKE - För helåren 1999 och 2000 - Underlag från experter, in. Stockholm: the Swedish National Board of Health and Welfare (Socialstyrelsen), 2002

147. Robinson MB: Acute regulation of sodium-dependent glutamate transport-ers: a focus on constitutive and regulated trafficking. Handb Exp Pharma-col:251-275, 2006

148. Robinson MB: The family of sodium-dependent glutamate transporters: a focus on the GLT-1/EAAT2 subtype. Neurochem Int 33:479-491, 1998

149. Romera C, Hurtado O, Botella SH, Lizasoain I, Cardenas A, Fernandez-Tome P, et al: In vitro ischemic tolerance involves upregulation of gluta-mate transport partly mediated by the TACE/ADAM17-tumor necrosis fac-tor-alpha pathway. J Neurosci 24:1350-1357, 2004

150. Ronne Engström E, Hillered L, Enblad P, Karlsson T: Cerebral interstitial levels of glutamate and glutamine after intravenous administration of nutri-tional amino acids in neurointensive care patients. Neurosci Lett 384:7-10, 2005

151. Ronne-Engström E, Cesarini KG, Enblad P, Hesselager G, Marklund N, Nilsson P, et al: Intracerebral microdialysis in neurointensive care: the use of urea as an endogenous reference compound. J Neurosurg 94:397-402, 2001

152. Ronne-Engström E, Hillered L, Flink R, Spännare B, Ungerstedt U, Carl-son H: Intracerebral microdialysis of extracellular amino acids in the hu-man epileptic focus. J Cereb Blood Flow Metab 12:873-876, 1992

153. Ronne-Engström ER, Hillered L, Flink R, Kihlström L, Lindquist C, Nie JX, et al: Extracellular amino acid levels measured with intracerebral mi-crodialysis in the model of posttraumatic epilepsy induced by intracortical iron injection. Epilepsy Res 43:135-144, 2001

154. Rothstein JD, Dykes-Hoberg M, Pardo CA, Bristol LA, Jin L, Kuncl RW, et al: Knockout of glutamate transporters reveals a major role for astroglial transport in excitotoxicity and clearance of glutamate. Neuron 16:675-686, 1996

155. Rothstein JD, Patel S, Regan MR, Haenggeli C, Huang YH, Bergles DE, et al: Beta-lactam antibiotics offer neuroprotection by increasing glutamate transporter expression. Nature 433:73-77, 2005

156. Ryttlefors M, Howells T, Nilsson P, Ronne-Engström E, Enblad P: Secon-dary insults in subarachnoid hemorrhage: occurrence and impact on out-come and clinical deterioration. Neurosurgery 61:704-714; discussion 714-705, 2007

68

157. Sarrafzadeh A, Haux D, Plotkin M, Ludemann L, Amthauer H, Unterberg A: Bedside microdialysis reflects dysfunction of cerebral energy metabo-lism in patients with aneurysmal subarachnoid hemorrhage as confirmed by 15 O-H2 O-PET and 18 F-FDG-PET. J Neuroradiol 32:348-351, 2005

158. Sarrafzadeh A, Haux D, Sakowitz O, Benndorf G, Herzog H, Kuechler I, et al: Acute focal neurological deficits in aneurysmal subarachnoid hemor-rhage: relation of clinical course, CT findings, and metabolite abnormalities monitored with bedside microdialysis. Stroke 34:1382-1388, 2003

159. Sarrafzadeh AS, Haux D, Ludemann L, Amthauer H, Plotkin M, Kuchler I, et al: Cerebral ischemia in aneurysmal subarachnoid hemorrhage: a correla-tive microdialysis-PET study. Stroke 35:638-643, 2004

160. Sarrafzadeh AS, Sakowitz OW, Kiening KL, Benndorf G, Lanksch WR, Unterberg AW: Bedside microdialysis: a tool to monitor cerebral metabo-lism in subarachnoid hemorrhage patients? Crit Care Med 30:1062-1070, 2002

161. Sarrafzadeh AS, Thomale UW, Haux D, Unterberg AW: Cerebral metabo-lism and intracranial hypertension in high grade aneurysmal subarachnoid haemorrhage patients. Acta Neurochir Suppl 95:89-92, 2005

162. Saveland H, Nilsson OG, Boris-Moller F, Wieloch T, Brandt L: Intracere-bral microdialysis of glutamate and aspartate in two vascular territories af-ter aneurysmal subarachnoid hemorrhage. Neurosurgery 38:12-19; discus-sion 19-20, 1996

163. Schierhout G, Roberts I: Anti-epileptic drugs for preventing seizures fol-lowing acute traumatic brain injury. Cochrane Database Syst Rev:CD000173, 2000

164. Schousboe A, Bak LK, Sickmann HM, Sonnewald U, Waagepetersen HS: Energy substrates to support glutamatergic and GABAergic synaptic func-tion: role of glycogen, glucose and lactate. Neurotox Res 12:263-268, 2007

165. Schulz MK, Wang LP, Tange M, Bjerre P: Cerebral microdialysis monitor-ing: determination of normal and ischemic cerebral metabolisms in patients with aneurysmal subarachnoid hemorrhage. J Neurosurg 93:808-814, 2000

166. Seki Y, Feustel PJ, Keller RW, Jr., Tranmer BI, Kimelberg HK: Inhibition of ischemia-induced glutamate release in rat striatum by dihydrokinate and an anion channel blocker. Stroke 30:433-440, 1999

167. Shen J, Petersen KF, Behar KL, Brown P, Nixon TW, Mason GF, et al: Determination of the rate of the glutamate/glutamine cycle in the human brain by in vivo 13C NMR. Proc Natl Acad Sci U S A 96:8235-8240, 1999

168. Skjoth-Rasmussen J, Schulz M, Kristensen SR, Bjerre P: Delayed neuro-logical deficits detected by an ischemic pattern in the extracellular cerebral metabolites in patients with aneurysmal subarachnoid hemorrhage. J Neu-rosurg 100:8-15, 2004

169. Smith JJ, Kampine JP: Circulatory Physiology - the essentials, ed 2nd. Baltimore: Lippincott, Williams and Wilkins, 1984

170. Springborg JB, Frederiksen HJ, Eskesen V, Olsen NV: Trends in monitor-ing patients with aneurysmal subarachnoid haemorrhage. Br J Anaesth 94:259-270, 2005

69

171. Stålhammar D, Starmark JE, Holmgren E, Eriksson N, Nordström CH, Fedders O, et al: Assessment of responsiveness in acute cerebral disorders. A multicentre study on the reaction level scale (RLS 85). Acta Neurochir (Wien) 90:73-80, 1988

172. Strauss DJ, Day SM, Shavelle RM, Wu YW: Remote symptomatic epi-lepsy: does seizure severity increase mortality? Neurology 60:395-399, 2003

173. Su TZ, Campbell GW, Oxender DL: Glutamine transport in cerebellar granule cells in culture. Brain Res 757:69-78, 1997

174. Suarez I, Bodega G, Fernandez B: Glutamine synthetase in brain: effect of ammonia. Neurochem Int 41:123-142, 2002

175. Suojaranta-Ylinen R, Kari A, Hernesniemi J, Vapalahti M, Takala J: Hy-permetabolism and increased peripheral release of amino acids after su-barachnoidal hemorrhage and its operative treatment. Nutrition 12:327-333, 1996

176. Szatkowski M, Barbour B, Attwell D: Non-vesicular release of glutamate from glial cells by reversed electrogenic glutamate uptake. Nature 348:443-446, 1990

177. Tagliaferri F, Compagnone C, Korsic M, Servadei F, Kraus J: A systematic review of brain injury epidemiology in Europe. Acta Neurochir (Wien) 148:255-268; discussion 268, 2006

178. Tanaka K, Watase K, Manabe T, Yamada K, Watanabe M, Takahashi K, et al: Epilepsy and exacerbation of brain injury in mice lacking the glutamate transporter GLT-1. Science 276:1699-1702, 1997

179. Tani H, Bandrowski AE, Parada I, Wynn M, Huguenard JR, Prince DA, et al: Modulation of epileptiform activity by glutamine and system A trans-port in a model of post-traumatic epilepsy. Neurobiol Dis 25:230-238, 2007

180. Teasdale GM, Drake CG, Hunt W, Kassell N, Sano K, Pertuiset B, et al: A universal subarachnoid hemorrhage scale: report of a committee of the World Federation of Neurosurgical Societies. J Neurol Neurosurg Psy-chiatry 51:1457, 1988

181. Thanki CM, Sugden D, Thomas AJ, Bradford HF: In vivo release from cerebral cortex of [14C]glutamate synthesized from [U-14C]glutamine. J Neurochem 41:611-617, 1983

182. Trotti D, Danbolt NC, Volterra A: Glutamate transporters are oxidant-vulnerable: a molecular link between oxidative and excitotoxic neurode-generation? Trends Pharmacol Sci 19:328-334, 1998

183. Trotti D, Rizzini BL, Rossi D, Haugeto O, Racagni G, Danbolt NC, et al: Neuronal and glial glutamate transporters possess an SH-based redox regu-latory mechanism. Eur J Neurosci 9:1236-1243, 1997

184. Ueda Y, Willmore LJ: Sequential changes in glutamate transporter protein levels during Fe(3+)-induced epileptogenesis. Epilepsy Res 39:201-209, 2000

185. Ungerstedt U: Microdialysis--principles and applications for studies in animals and man. J Intern Med 230:365-373, 1991

186. van der Worp HB, Thomas CE, Kappelle LJ, Hoffman WP, de Wildt DJ, Bar PR: Inhibition of iron-dependent and ischemia-induced brain damage

70

by the alpha-tocopherol analogue MDL 74,722. Exp Neurol 155:103-108, 1999

187. Van Hemelrijck A, Sarre S, Smolders I, Michotte Y: Determination of amino acids associated with cerebral ischaemia in rat brain microdialysates using narrowbore liquid chromatography and fluorescence detection. J Neurosci Methods 144:63-71, 2005

188. Vespa P, Bergsneider M, Hattori N, Wu HM, Huang SC, Martin NA, et al: Metabolic crisis without brain ischemia is common after traumatic brain in-jury: a combined microdialysis and positron emission tomography study. J Cereb Blood Flow Metab 25:763-774, 2005

189. Vespa P, Prins M, Ronne-Engström E, Caron M, Shalmon E, Hovda DA, et al: Increase in extracellular glutamate caused by reduced cerebral perfusion pressure and seizures after human traumatic brain injury: a microdialysis study. J Neurosurg 89:971-982, 1998

190. Vespa PM, Nuwer MR, Nenov V, Ronne-Engström E, Hovda DA, Bergs-neider M, et al: Increased incidence and impact of nonconvulsive and con-vulsive seizures after traumatic brain injury as detected by continuous elec-troencephalographic monitoring. J Neurosurg 91:750-760, 1999

191. Vespa PM, O'Phelan K, McArthur D, Miller C, Eliseo M, Hirt D, et al: Pericontusional brain tissue exhibits persistent elevation of lactate/pyruvate ratio independent of cerebral perfusion pressure. Crit Care Med 35:1153-1160, 2007

192. Voutsinos-Porche B, Bonvento G, Tanaka K, Steiner P, Welker E, Chatton JY, et al: Glial glutamate transporters mediate a functional metabolic crosstalk between neurons and astrocytes in the mouse developing cortex. Neuron 37:275-286, 2003

193. Waagepetersen HS, Qu H, Hertz L, Sonnewald U, Schousboe A: Demon-stration of pyruvate recycling in primary cultures of neocortical astrocytes but not in neurons. Neurochem Res 27:1431-1437, 2002

194. Weir B, Findlay J: Subarachnoidal hemorrhage, in Carter LP, Spetzler RF (eds): Neurovascular Surgery. New York: McGraw-Hill, 1995, pp 557-581

195. Wester P, Asplund K, Eriksson S, Holm J, Marké L, Norlund A, et al: Stroke, in. Stockholm: SBU-The Swedish Council on Technology Assess-ment in Health Care, 1992

196. Willmore LJ: Post-traumatic epilepsy: cellular mechanisms and implica-tions for treatment. Epilepsia 31 Suppl 3:S67-73, 1990

197. Willmore LJ, Rubin JJ: Antiperoxidant pretreatment and iron-induced epi-leptiform discharges in the rat: EEG and histopathologic studies. Neurol-ogy 31:63-69, 1981

198. Willmore LJ, Rubin JJ: Effects of antiperoxidants on FeCl2-induced lipid peroxidation and focal edema in rat brain. Exp Neurol 83:62-70, 1984

199. Willmore LJ, Rubin JJ: Formation of malonaldehyde and focal brain edema induced by subpial injection of FeCl2 into rat isocortex. Brain Res 246:113-119, 1982

200. Winkler BS, Kapousta-Bruneau N, Arnold MJ, Green DG: Effects of inhib-iting glutamine synthetase and blocking glutamate uptake on b-wave gen-eration in the isolated rat retina. Vis Neurosci 16:345-353, 1999

71

201. Xie M, Wang W, Kimelberg HK, Zhou M: Oxygen and glucose depriva-tion-induced changes in astrocyte membrane potential and their underlying mechanisms in acute rat hippocampal slices. J Cereb Blood Flow Metab, 2007

202. Xu GY, McAdoo DJ, Hughes MG, Robak G, de Castro R, Jr.: Considera-tions in the determination by microdialysis of resting extracellular amino acid concentrations and release upon spinal cord injury. Neuroscience 86:1011-1021, 1998

203. Zauner A, Bullock R, Kuta AJ, Woodward J, Young HF: Glutamate release and cerebral blood flow after severe human head injury. Acta Neurochir Suppl (Wien) 67:40-44, 1996

204. Zhang M, Li WB, Geng JX, Li QJ, Sun XC, Xian XH, et al: The upregula-tion of glial glutamate transporter-1 participates in the induction of brain is-chemic tolerance in rats. J Cereb Blood Flow Metab 27:1352-1368, 2007

205. Zschocke J, Bayatti N, Clement AM, Witan H, Figiel M, Engele J, et al: Differential promotion of glutamate transporter expression and function by glucocorticoids in astrocytes from various brain regions. J Biol Chem 280:34924-34932, 2005

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 329

Editor: The Dean of the Faculty of Medicine

A doctoral dissertation from the Faculty of Medicine, UppsalaUniversity, is usually a summary of a number of papers. A fewcopies of the complete dissertation are kept at major Swedishresearch libraries, while the summary alone is distributedinternationally through the series Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty ofMedicine. (Prior to January, 2005, the series was publishedunder the title “Comprehensive Summaries of UppsalaDissertations from the Faculty of Medicine”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-8630

ACTA

UNIVERSITATIS

UPSALIENSIS

UPPSALA

2008

glutamate turnover and energy metabolism in brain injury

Documents