THE EFFECTS OF VOLATILE AND INJECTABLE ANESTHETICS ON THE NEUROCHEMICAL PROFILE IN RAT BRAIN: STUDIE

1
THE EFFECTS OF VOLATILE AND INJECTABLE
ANESTHETICS ON THE NEUROCHEMICAL PROFILE IN RAT
BRAIN STUDIES WITH



HIGH-RESOLUTION MAGIC ANGLE SPINNING 1H
MAGNETIC RESONANCE SPECTROSCOPY (HR-MAS-1H-MRS)
AT 11.7T K. Murphy A. Pustavoitau G.M.
McKelvey H.M. Marsh J. Tang N.
Seraji-Bozorgzad G.J. Moore M.P. Galloway.
Departments of Anesthesiology and Psychiatry and
Behavioral Neuroscience, Wayne State University
School of Medicine, Detroit, MI, USA.
Fig 1
Results
Introduction
Summary
Table 1
Selected 1H-MR Visible Neurochemicals (Rat
striatum 4mg)
The site(s) of action of general anesthetic
agents within the central nervous system remain
largely speculative. However, the advent of
proton magnetic resonance spectroscopy (1H-MRS)
provides a non-invasive method to determine
drug-induced changes in levels of
neurotransmitters (GABA, glutamate), precursors
(glutamine), and metabolites in vivo. Given the
unique nature of the measurement derived from MR
analysis (i.e. in intact tissue), combined with
the inability to differentiate metabolic from
neuronal pools, the precise neurobiological
significance of changes in MR-visible
neurochemicals remains to be determined.
Therefore, the objectives of the current studies
were to determine 1) the effect of potentiating
GABA-A receptor mediated inhibition with either
propofol or gaseous anesthetics on the
neurochemical profile of MR-visible
neurochemicals, and 2) the regional profile of
anesthetic action in rat brain. The metabolic
profile of MR-visible neurochemicals was
determined with high resolution magic angle
spinning (HR-MAS) 1H-MRS at 11.7T, which is a
novel neuropharmacological application of 1H-MRS
that provides a highly resolved proton chemical
shift spectrum. Although clinical 1H-MRS is a
powerful (and somewhat routine) technique, 1H
-MRS spectra obtained in vivo have limited peak
resolution due to relatively low field strengths.
For example, interference from various undefined
macromolecules in the tissue voxel under study
leads to further line-broadening and distortion
of the baseline.  While increased field strength
helps to improve spectral resolution, the problem
of macromolecule interference remains a hindrance
to high resolution spectroscopy, especially in
semi-solid samples, where dipole-dipole
interactions are not averaged out by the tumbling
of molecules (as they are in liquids). These
interactions are minimized when the sample is
spun around an axis at an angle (magic angle)
to the static magnetic field to simulate the
molecular tumbling present in a liquid sample.
By determining the effect of anesthetics on the
highly-resolved spectrum generated with HR-MAS
1H-MRS, we envision development of a
hypothesis-driven paradigm that can be directly
translated to clinical research studies.
Effects of Propofol, Isoflurane and Halothane on
the Metaboloic Profile of 1H-MRS Visible
Neurochemicals

• Acute exposure to either volatile or injectable
  anesthetics produced distinctive alterations in
  the metabolomic profile of those neurochemicals
  amenable to analysis with proton magnetic
  resonance spectroscopy.
• Although each drug was accompanied by the loss of
  righting reflex as well as analgesia, the effects
  on the neurochemical profile as well as the
  regional profile distinguished propofol from the
  volatile anesthetics.
• Decreased levels of GLUMRS after propofol or
  isoflurane are consistent with potentiation of
  endogenous GABA at inhibitory GABA-A synapses on
  glutamatergic pyramidal neurons in both the
  midbrain and cortex. Similarly, decreased GLU/GLN
  ratios are consistent with diminished glutamate
  turnover in the neuronal-glial metabolic shuttle.
  Blockade of glutamatergic drive with a sedating
  dose (100-300 mg/kg iv) of the NMDA receptor
  antagonist ketamine did not decrease GLUMRS
  (data not shown).
• Increased DA levels in striatal projection fields
  after propofol or volatile anesthetics are
  consistent with potentiated GABA-ergic inhibition
  of DA unit activity.
• Hypnotic agents selectively decreased levels of
  GABAMRS suggesting a complex interaction with
  interneurons.
• Increased LACMRS after hypnotic agents is
  consistent with compromised metabolic energy
  status.
• The regional pattern of changes in GLU after
  propofol or isoflurane is consistent with a
  GABA-mediated inhibition of cortical-thalamic
  pyramidal neurons, as well as their reciprocal
  connections.

CRE
NAA
BET
LAC
GPC
CRE
TAU
PCH
GLU
GLN
GLU
CHO
GLN
INS
GABA
PEA
NAA
GABA
INS
GABA
GSH
ALA
LAC
NAA
SUC
In each instance with statistically significant
drug effects, the top number refers to the
absolute concentration (nmol/mg tissue) in the
treated tissue. The bottom number refers to the
drug effect as percent of control.
p lt 0.05 significantly lower compared to controls
Fig 2
p lt 0.05 significantly higher compared to controls
Selected 1H-MR Visible Neurochemicals
Fig 5
Propofol decreased Gluamate/Glutamine ratios in
selected brain regions

1H-MRS Gluamate/Glutamine ratios in selected
brain regions showed consistently lower ratios in
propofol treated animals compared to those of
controls. Decreased GLU/GLN ratios are
consistent with diminished glutamate turnover in
the neuronal-glial metabolic shuttle. Each point
is the mean /- SEM for N5 subjects p lt 0.05
vs control.
Discussion



The primary conclusion from these studies is that
anesthetic-induced potentiation of GABA-A
receptor activity decreases glutamate levels
determined with proton magnetic resonance
spectroscopy. Moreover, the neuroanatomy of the
GLU effect suggests that anesthetics act on
principal neurons in both the cortex and
thalamus. Based on these observations, we
predict that clinical studies with a 4 Tesla
magnet would reveal decreased GLUMRS (or GLx at
lower field strengths) in the cingulate cortex of
subjects anesthetized with either propofol or
isoflurane. Decreased pyramidal neuronal activity
may arise from either decreased afferent
stimulation or activation of interneurons,
representing indirect or direct enhancement of
GABA-ergic tone respectively. However the
mechanism by which GABA-A inhibition of pyramidal
neurons decreases GLU remains to be determined.
Glutamate levels are maintained primarily by
phosphate-activated-glutaminase (PAG) action on
neuronal glutamine or transaminase action on
a-ketoglutarate derived from the tricarboxylic
acid cycle. Additionally, the availability of the
precursor glutamine, synthesized in glia by
glutamine synthetase, may contribute to
regulation of GLU levels. Thus, it is
conceivable that pyramidal cell hyperpolarization
may 1) diminish glutamine transport into the
principal neurons and/or 2) decrease
mitochondrial PAG activity by acute
post-translational protein modification such as
phosphorylation. The differential anatomy of the
GLU response to propofol and the volatiles may
reflect the diversity of GABA-A receptors.
Specific subunits of the pentameric GABA-A
receptor confer sensitivity to the behavioral
effects (viz. sedation, anxiolysis, hypnosis,
amnesia, analgesia, immobility, et al) of
benzodiazepines, ethanol, and anesthetics.
Therefore, the absence of an effect on GLUMRS
may suggest the absence of anesthetic-sensitive
GABA-A receptors on principal GLU cells in a
particular region of interest (e.g. hippocampus
CA1, striatal subregions). Besides differential
subunit composition, GABA-A receptor strength is
sensitive to regional differences in receptor
location (synaptic phasic or perisynaptic tonic),
receptor density, phosphorylation, extracellular
GABA, and presynaptic GABA-B autoreceptors. The
ability of isoflurane, but not halothane, to
decrease GLUMRS in either the prefrontal or
cingulate cortex is consistent with the
observation that the efficacy of halothane (1.2)
to decrease rat cortical unit activity is
significantly less than isoflurane (1.1) when
measured in intact animals in vivo (Hentschke et
al 2005). In addition to GABA-A receptor
diversity, modulation of GABA or GLU inputs by
coincident dopamine innervation may affect the
GLUMRS response. For example, excitatory
projections from the periventricular nucleus
(PVN) of the thalamus converge with DA neurons
onto common targets on spines of projection
neurons in the nucleus accumbens whereas the PVN
projections to the prefrontal cortex shows a
minimal degree of convergence with mesocortical
DA terminals (Pinto et al 2003). In the present
case where DA modulation is suspended (see
below), dysregulated thalamo-accumbal excitatory
afferents may counteract the propofol-enhanced
activity at GABA-A receptors. Increased dopamine
levels in striatal projection fields after
propofol or volatile anesthetics is consistent
with enhanced activity at GABA-A receptors and
subsequent attenuation of DA neuronal unit
activity. Midbrain DA neurons express GABA-A
subunits and are innervated by GABA afferents
from the striatum, globus pallidus, and
substantia nigra reticulata. It is well
established that treatment with the GABA
precursor ?-butyrolactone (GBL) inhibits DA
neuronal activity with a resultant increase in DA
levels (an effect that is blocked by
administration of DA autoreceptor agonists to
suppress DA synthesis). The most prominent
GABA-A subunit arrangement in rat DA cells is the
a1ß2?2 composition conferring sensitivity to
benzodiazepines additionally ß3 subunits (which
confer propofol anti-nociceptive sensitivity)
have been detected in DA cells. In summary,
this is the first report of anesthetic-induced
alterations in the rat neurochemical profile
determined with high-field high-resolution
1H-MRS. Changes in the neurochemical profile of
glutamate, glutamine, GABA, lactate, et al. is
interpreted in the framework of current theories
of molecular and anatomical actions of propofol
and volatile anesthetics. Given the close
identity of the chemical shift spectra determined
in the present studies with that determined in
clinical 1H-MRS, the results are directly
translatable to clinical studies on the
neuropharmacology of anesthesia.



Fig 3

Representative Regions of Interest 2.1
mm punches

Methods
Propofol treatment Male Sprague Dawley rats
(225-250 grams) were catheterized in the tail
vein with 24 gauge pediatric catheters that were
secured to the tail.  Treated animals (n6)
received an anesthetic dose of intravenous
propofol (15 mg/kg q.15 min x5, 250 µL bolus) and
were sustained in the anesthetic state for 60
min.  Anesthetic state was defined as loss of the
righting reflex, immobility, and unresponsiveness
to corneal, auditory or pain stimuli. Volatile
treatment Three groups (n5) of male
Sprague-Dawley rats (300-340 grams) were exposed
to 1) room air, 2) isoflurane (1.1 vol), or 3)
halothane (1.2 vol) at 100 O2 for 1 hour.  A
Datex-Ohmeda Modulus II anesthesia machine
(Datex-Ohmeda Inc. Madison, WI) delivered
anesthetic gases and O2 to a custom-made
plexiglass chamber with a 40L volume. Gas
concentrations were measured and monitored with
an RGM Ohmeda 5250 (American Lab Medical
Equipment, Inc, FL) and maintained at constant
levels (1 MAC) O2 and CO2 levels were
continuously monitored for physiological
stability. Animals were sacrificed by
decapitation after approximately 60 minutes of
exposure to the volatiles (1MAC-hour). After
drug treatment, animals were decapitated, brains
rapidly removed, and 2 mm coronal slices were
obtained with the aid of a chilled rat brain
matrix.  Punches (2.1 mm diameter) were taken
from slices (see figure 4), immediately frozen on
solid CO2, and stored at -80C until HR-MAS 1H-MRS
analysis. HR-MAS 1H-MRS of the intact tissue
punches (4 mg) was performed with standard CPMG
techniques using a Bruker Avance 500 11.7T
magnet. Absolute quantification of each
neurochemical (nmol/mg tissue) was determined
with a custom-designed LCModel (Provencher, 1993)
utilizing a linear combination of 27
neurochemical model spectra (basis set) to fit
the tissue spectrum and calculate absolute
concentration values for each neurochemical.  The
basis set was custom-made using calibrated
phantoms of individual neurochemicals in buffer.
The goodness of fit to each spectra was assessed
statistically with Cramer-Rao bounds and a CR
value lt 25 was required for further analysis. 
Absolute concentrations of MR visible metabolites
were corrected for tissue weight and are
expressed as nmol/mg of wet weight.  Statistical
significance between group means was determined
with an appropriate t-test (2 tail,
Plt0.05). Monoamine levels were determined in
identical tissue punches contralateral to the
sample used for 1H-MRS. Tissues were sonically
disrupted in 0.4N perchloric acid, centrifuged,
and an aliquot injected directly on a C18 column
and detected with an ESA coulometric detector.
Fig 6
Propofol Increases Striatal Dopamine and
Metabolites
Enhanced GABA-ergic tone after propofol (15 mg/kg
iv q 15 min x5) increased striatal dopamine
levels as well metabolites. DA in the N accumbens
was also increased (23, Plt0.05 data not shown).
The DA increase may reflect ongoing tyrosine
hydroxylation (ie DA synthesis) in the absence of
impulse-dependent DA release, subsequent to
GABA-A mediated inhibition of DA cell firing. The
increase in DA is similar to that observed after
GBL (gamma-butyrolactone, a GABA precursor),
which is thought to act at GABA-B receptors to
hyperpolarize DA neurons. Each point is the mean
/- SEM for N5 subjects P lt 0.05 vs control.
Discussion
Fig 4
Neurophysiologic Theory of the Action of
Anesthetics to Suppress Awareness
(6)
(5)
Cortex- Pre Frontal Cortex
Cellular pathways of Neuronal and Astrocyte
function
Fig 7
N. Reticularis
(4)
The anesthetic-induced potentiation of GABA-A
receptor activity decreases glutamate levels when
determined with proton magnetic resonance
spectroscopy. Moreover, the neuroanatomy of the
GLU effect suggests that anesthetics act on
principal neurons in both the cortex and thalamus
Decreased pyramidal neuronal activity may arise
from either decreased afferent stimulation or
activation of interneurons, representing indirect
or direct enhancement of GABA-ergic tone
respectively. The availability of the precursor
glutamine, synthesized in glial by glutamine
synthetase, may contribute to regulation of GLU
levels. Thus, it is conceivable that pyramidal
cell hyperpolarization may 1) diminish glutamine
transport into the principal neurons and/or 2)
decrease mitochondrial PAG activity by acute
post-translational protein modification such as
phosphorylation.
GABA
Anesthetic Exogenous - multi modal sensory input
(3)
Limbic system (Cingulate)
Thalamus
(1)
(2)
ARAS
Neurophysiologic Theory of the Action of
Anesthetics to Suppress Awareness Step 1
Depression of the brainstem reduces the
influences of the Ascending reticicular
activating system (ARAS) on the thalamus and
cortex. Step 2 Depression of mesolimbic-dorsolat
eral (cingulate) - prefrontal cortex interactions
leads to blockade of memory storage. Step 3
Further depression of the ARAS releases its
inhibition of the nucleus reticularis of the
thalamus, resulting in closure of thalamic gates
by hyperpolarizing GABA-mediated inhibitory
action of the nucleus blocking Step 4
Thalamocortico reverberations and perception so
that Step 5 Parietal-frontal transactions are
uncoupled), blocking cognition, and Step 6
Prefrontal cortex is depressed to reduce
awareness Adapted from John and Prichep,
Anesthesiology (2005) 102 (2) 447-469
References John ER and Prichep LS, Anesthesiology
102 (2) 447-469, 2005 Hentschke H et al. Eur J
Neuroscience, Vol. (21) 93102, 2005 Pinto A et
al. J Comp Neurol. 28459(2)142-55,
2003 Acknowledgements Supported by NIDA
R01-DA-16736 Special thanks Kristen Haubenreich
and Shonagh OLeary Moore for HPLC analysis