tration following fluid percussion head injury in rats. We report that injury caused a profound and rapid decrease in intracellular free Mg2+ which was signifi-.
Vol. 263, No. 2, Issue of January 15, pp.,757-761,1988 Prrnted in U.S.A.
JOURNALOF BIOLOGICAL CHEMISTRY 1988 by The American Society for Biochemistry and Molecular Biology, Inc.
THE Q
Decline in Intracellular Free Mg2+Is Associated with Irreversible Tissue Injury after BrainTrauma* (Received for publication, July 24, 1987)
Robert Vink$J, Tracy K. McIntosh$, Paul DemedlukS, MichaelW. Weinerq, and AlanI. Faden$ From the $Center for Neural Injury andIIMagnetic Resonance Unit, Departments of Neurology, Medicine and Radiology, University of California, San Francisco 94143 and the Veterans Administration Medical Center, San Francisco, California94121
Much of the tissue damage resulting from trauma to the central nervous system appears to result from secondary, delayed biochemical changes that follow primary mechanical injury. However, the early biochemical events remain to be elucidated. In the present studies, we have used phosphorus ("'P) magnetic resonance spectroscopy (MRS) toexamine in uiuo, the temporal changes in brain intracellular free Mg2+ concentration following fluid percussion head injury in rats. We report that injury caused aprofoundand rapid decrease in intracellular free Mg2+which was significantly correlated with the severity of injury. At high levels of injury, the decrease in intracellular free Mg2+ concentration was associated with a decrease in total Mg2+ concentration as determined by atomic absorption spectrophotometry. Prophylactic treatment with MgSO, prevented the post-traumatic decrease in intracellular free Mg2+ and resulted in a significant improvement in acute neurological outcome. Because magnesium is essential for a number ofcritical enzyme reactions, including those of glycolysis, oxidative and substrate level phosphorylation, protein synthesis, and phospholipid synthesis, changes in free Mg2+ after brain traumamay represent a critical early factor leading to irreversible tissue damage.
The mechanisms by which central nervous system trauma leads to irreversible tissue damage are poorly understood. Secondary biochemical changes including, among others, derangement of mitochondrial energy production, release of excitatory amino acids, production of oxygen-free radicals, release of endogenous opioids, and activation of calcium cascadeeffects (1-6) are believed to contribute to the injury process. An important limitation in studiesof brain or spinal cord trauma has been the dependence upon single time point, invasive techniques to assess the biochemical changes that follow mechanical injury. The application of magnetic resonance spectroscopy (MRS)' to the studyof brain metabolism (7) now permits non-invasive, repeated measurement of cer* This work was supported in part by Center for Disease Control United States Public Health Service Grant 902269, United States Army Medical Research and Development Command Contract 1785-PP-5843 (to A. I. F. and T. K. M.), and by a Veterans Administration Merit Review (to T. K. M.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 8 Recipient of a Sandoz Neuroscience Fellowship. To whom correspondence should be addressed Neurology Service (127), Veterans Administration Medical Center, 4150 Clement St., SanFrancisco, CA 94121. The abbreviation used is: MRS, magnetic resonance spectroscopy.
tain biochemical events following experimental trauma(8,9). In addition to showing changes in high energy phosphates and intracellular pH, phosphorus ("P) spectroscopy allows themeasurement of intracellular free Mgz' concentration (10). Mg2+ is essential for a number of critical enzymatic reactions, including those of glycolysis and oxidative phosphorylation (11,12). Furthermore, optimiumMg' concentrations are required for DNA transcription and protein synthesis (13), as well as a variety of plasma membrane functions (14). In the present studies, we have utilized "P MRS to follow dynamic changes in free intracellular M$+ after fluid percussion-induced brain trauma in rats. MATERIALS AND METHODS
Animal Preparation and MRS Data Acquisition-Male, SpragueDawley rats (350-400 g) were anesthetized with 80 mg/kg ketamine and 35 mg/kg sodium pentobarbital and prepared for MRS studies as previously described (8, 9). The dorsal surface of the skull was exposed, skin and temporal muscles retracted, and a plastic trauma screw inserted into a 2-mm craniotomy centered over the left hemisphere. Atwo-turn 5 X 9-mm MRS surface coilwas positioned centrally around the traumasite, and 31PMRS spectra were obtained in 20-min blocks on a GE CSI 2T MRS spectrometer using a 90" pulse and a 0.7-s repetition rate prior to and following brain injury. Brain Injury-Brain injury was induced using a fluid percussion model (5, 6) in which a bolus of37 "C isotonic saline is rapidly injected into the closed cranial cavity, causing transient brain deformation. The degree of trauma was regulated so as to produce either low (0.9 & 0.4 atm; n = 7) or high (2.0 + 0.5 atm; n = 10) injury as determined by cardiovascular, electrophysiologic, and neurological changes (5, 6, 8, 9). Mean arterial blood pressure was continuously monitored, and arterial blood pH, PO,, pCO,, and HCO, were repeatedly analyzed throughout the procedure. A parallel group of animals ( n = 10 for both injury levels) were injured in an identical fashion outside of the MRS spectrometer, and EEG recording utilizing Fast Fourier transformed, compressed spectral array technology (Siegen neuroscope) was obtained in addition to the physiological variables described above. Free M$* Determinations-The intracellular free M$+ concentration was calculated from the chemical shift of the @-ATPpeak relative to thea peak of ATP as described by Gupta et al. (10).For any given pH, the proportion ofMgZ'-bound ATP relative to M e - f r e e ATP determines the 31PMRS chemical shift of ATP (15,16) in such a way that the relative concentration of Mg2"free ATP to total ATP can be determined according to the following equation (17):
4 = [ATPtre]/[ATPt.,t.~] = (a$'," - 6Y#ATP)/(6$TP
- G?gATP)
(I)
where 6$' is the chemical shift difference (in hertz) between the aATP and the 8-ATP in the sample spectrum, 6 y is the chemical shift difference between the a-ATP and the @-ATPof 100% M P bound ATP, and 6 s ' is the chemical shift difference between the aATP and the @-ATPof M p - f r e e ATP. In vitro studies have shown that atphysiological pH the 6YfATP is 8.36 ppm and';6: is 10.82 ppm (10, 18-20;see also Table I). From the 4 and a knowledge of the Mg2"ATP dissociation constant, free MgZ' can thus be
757
Free M$+ Decline following Trauma Brain
758
determined from the following equation (17): [Mg2+]= Icf;'"""
(@-I
- 1)
(2)
is assumed where at 37 "C, pH 7.2, and 0.15 M ionic strength, to be50 p M (18, 21). Presumably, any change in free Mg2' is a reflection of both the mitochondrial and cytosolic pools, each of which has essentially the same free Mg2+ concentration (22). Intracellular pH was determined according to the following equation (7): pH = 6.77
+ lOg(6Pi - 3.29)/(5,68 - 6Pi)
(3)
where 6Pi is the chemical shift of the Pi relative to phosphocreatine which was used as an internal standard. Total M$+ Determinations-Total brain M$+ was determined in high injury and control (non-injured) animals( n= 6/group) by atomic absorption spectrophotometry. Xigh injury was induced in a manner identical to that described for the MRS studies. At 1-h post-injury, thebrain was frozen in situ with liquid nitrogen, bisected, and removed while still frozen. Injured hemisphere brain tissue was then extracted and assayed for Mg2+ content as previously described (23). Mg2+ Treatment-Animals ( n= 5/group) were infused with either MgSO, (15 medliter) or saline (equal volume) over a 5-min period immediately prior to high injury. This rateof MgSO, infusion resulted in a slight but non-significant fall in mean arterial blood pressure (mean decrease = 7 mm Hg), that was transient and had returned to normal by the time injury was induced. In Vitro Studies-To determine the effects of M e binding on the chemical shift of ATP, an ATPsolution made up of 20 mM ATP, 75 mM KCI, 75 mM NaCI, 10 mM Pi, pH 7.1, was titrated with MgSO,. The chemical shift of ATP was determined to within 0.08 ppm. The effects of other ions on the chemical shift of the MgATP complex under physiological conditions was determined on a solution made up of 20 mM ATP, 50 mM KCl, 10 mM Pi, and 17 mM MgSO,, pH 7.1. The concentrations of the individual components were chosen so as toimitate the K+ concentration and theM e - b o u n d state of ATP (and therefore the free MgZ+ concentration) found in rat brain in vivo prior to trauma (Table I). Na+, Kt,and Ca2+were added to this solution in concentrations which would reflect possible changes in the intracellular ionic ratios in uivo following fluid percussion injury (3). Neurological Scoring-All animals were scored by blinded observers
TABLE I Effects of Mgz+, Ca2+,K+, and NQ' concentration on the chemical shift difference between a- and @-ATP( 6 3 CIS determined by 31P
MRS pH was maintained at 7.10 f 0.05 by addition of either HCl or NaOH. The effects of M%+ concentration were determined by titration of 20 mM ATP, 75 mM KCI, 75 mM NaCl, 10 mM P,, pH 7.10, with MgSO,. A concentrated ATP solution was used to enhance signal to noise in the 31PMRS spectra and reduce error in chemical shift determinations. To determine the effects of post-traumatic fluctuations in Ca2+, Na', and K' ratioson the Me-dependent chemical shift difference between a- and @-ATP(ae8), the respective cations were added to a solution containing 20 mM ATP, 10 mM Pi, 50 mM KC1, 1 7 mM MgSO,, pH 7.1. The 6, of this solution was identical to thatfound by 31PMRS in rat brain in uivo prior to injury, giving a free M e concentration in this solution of 1.0 mM. 6,s
Additions
20 mM ATP, 75 mM mM KC1, 75 10 mM Pi
9.31 8.52 8.36
(k0.08 ppm)
NaCI, 10.82
+5 mM +10 mM MgS04 +15 mM MgSO, +20 mM MgS04 +25 mM MgSO, +30 mM MgSO4
9.94
20 mM ATP, 50 mM KCl, 10 mM Pi, 17 mM MgSO, +BO mM NaCl +lo mM CaC12 +25 mMKC1 +40 mM EDTA
8.49
8.43 8.36
8.54 8.54 8.54 10.38
for post-traumatic neurological deficit at 24 h post-injury. Neurological outcome measures included: ( a ) forelimb flexion uponsuspension by the tail; ( b ) decreased resistance to lateral pulsion; (c) circling behavior upon spontaneous ambulation; (d) ability to stand on an inclined angle board with the maximum angle at which the animal can stand for 5 s recorded (angle board); and ( e ) the latency to traverse a narrow (2 cm wide) wooden balance (beam traverse). All animals were graded as follows for each task 4, normal; 3,mild deficit; 2, moderate deficit; 1, severe deficit; 0, afunctional. A composite neuroscore was developed for each animal by combining the scores of tests a-e so that: 20, normal; 15, slightly impaired; 10, moderately impaired; 5, severely impaired: 0, afunctional. Data Analysis-Neurological scores were evaluated utilizing nonparametric Kruskal-Wallis analysis of variance (ANOVA)followed by individual non-parametric Mann-Whitney U tests; all other data was analyzed using ANOVA followed byindividual Student NewmanKeuls tests, and by regression analysis. A p value
