Stephen J Sawiak1, Bianca Jupp1, Tom Taylor1, Daniele Caprioli1, T Adrian Carpenter1, and Jeffrey Dalley1
1University of Cambridge, Cambridge, United Kingdom
Synopsis
Disorders of impulse control are a rising
issue in society as diagnosis rates of conditions such as attention deficit and
hyperactivity disorder are increasing. In humans, the MEGAPRESS approach to
measuring GABA is becoming a standard technique but it has not yet been used
much in translational studies. Here, we used it to measure GABA in the striatum
of highly-impulsive rats compared to rats with low impulsivity and found
significantly reduced levels of this inhibitory neurotransmitter in the
impulsive animals. Purpose
γ-aminobutyric acid (GABA) is the chief
inhibitory neurotransmitter in the brain and its disturbance is related to a
range of neuropsychiatric conditions. Accurate in vivo measurements of GABA are
hampered by its overlap with the much more abundant metabolites glutamate,
glutamine and creatine. Mescher-Garwood PRESS (MEGA-PRESS), a spectral editing
technique, has been widely applied to unmask GABA signals in humans but has seen
little application to preclinical models
1.
Here, we used MEGAPRESS to determine whether GABA levels differ in the striatum
of rats selected for extreme levels of impulsivity with an automated
behavioural task. These rats can be used to model a range of disorders of
impulse control, including ADHD.
Methods
Behavioural screening
Adult
male rats (Lister-hooded, n = 96) were screened for impulsivity using the
five-choice serial reaction time task (5-CSRTT)2. Briefly, in each trial rats must correctly
respond to one of five randomly illuminated recesses in a chamber. A correct
response leads to a sugar pellet reward whereas an incorrect response,
premature response, or failure to respond leads to darkening of the chamber for
five seconds. A new trial occurs every five seconds until 100 trials are
complete. After four months of training, rats are tested repeatedly over a
three week period with a longer inter-trial interval (seven seconds) as a
challenge to reveal impulsive behaviour. Impulsivity here was quantified by the
rate of premature responses in successive long inter-trial interval tests. Of
the 96 initial rats, the 12 rats with the lowest rates of premature responses
formed the low-impulsive (LI) group and the 12 rats with the highest rates
formed the high-impulsive (HI) group.
Image and spectroscopic acquisition
Rats
were anaesthetised with isoflurane (1-3% in 2l/min O2) with
respiration rates used to monitor depth and adjust the dose as necessary. A
Bruker BioSpec 47/40 4.7T system was used with ParaVision 5.1 (Bruker,
Germany). A four-channel rat brain array was used for signal reception. For
localisation, 250µm isotropic images were obtained (RARE: TR/TE 15.3s/36ms,
256×256 matrix, 64×64mm2 field of view) and the
manufacturer-provided MAPSHIM was used for field map image based shimming.
Voxels were placed at the level of the inferior ventral striatum encompassing
both hemispheres (3×7×4mm3) and MEGA-PRESS was acquired following ref. 3 (TR/TE 2000/68ms; alternating refocussing
pulses of 20ms at 1.9ppm and 7.5ppm for 256 transients of 2,048 points at 4kHz
bandwidth) with a total acquisition time of 8.5 min per animal.
Processing
Data were zero-filled and line broadened by 3Hz. Each transient was frequency
and phase corrected by fitting a Lorentzian curve to the creatine peak at
3.0ppm, discarding transients where these differed by more than 4 standard
deviations from the median value. The n-acetyl aspartate (NAA) peak was used as
a reference as it was the strongest signal in each spectrum. Peak areas of each
were evaluated by fitting Lorentzian peaks using Matlab (Mathworks, Inc.).
Results
Voxel placement for a typical subject is
shown in figure 1, with exemplar spectra shown in figure 2 for MEGA-PRESS ‘on’
and ‘off’. Typical frequency drift was 3Hz and on average 13 transients were
discarded from each series according to the rejection criteria.
The stability of premature responses in LI and
HI rat groups across successive short (SITI1, SITI2,..) and long interval challenge
(LITI1, LITI2) 5CSRTT trials is shown in Figure 3A, with GABA/NAA ratios shown for HI
and LI groups in figure 3B. Across all rats, the mean [GABA]/[NAA] ratio was
0.22±0.08. HI rats had, on average, 28% lower GABA/NAA ratios (p = 0.02;
one-tailed Student’s t-test). To
mitigate the possibility that aspartate changes led to the decreased ratios, we
calculated [GABA]/[creatine] and [GABA]/[choline] ratios which were reduced by
23% and 26% in the high impulsive rats, respectively.
Discussion and Conclusion
For the first time we have shown that in
vivo GABA ratios are disrupted in rats exhibiting extreme high/low impulsivity
which may prove to be an important neural endophenotype in disorders of impulse
control. We have shown that MEGAPRESS measurements of GABA ratios are
not only feasible in rats in a reasonable time at the relatively low field
strength of 4.7T but useful for demonstrating differences in a translational
model. These findings add to growing evidence of a role for impaired GABA functionality
in neuropsychiatric disorders of impulse control
4,5.
Acknowledgements
No acknowledgement found.References
1. Mullins, P.G., et al., Current practice in the use of MEGA-PRESS
spectroscopy for the detection of GABA. Neuroimage, 2014. 86: p. 43-52.
2. Robbins, T.W., The 5-choice serial reaction time task:
behavioural pharmacology and functional neurochemistry. Psychopharmacology
(Berl), 2002. 163(3-4): p. 362-80.
3. Mescher, M., et al., Simultaneous in vivo spectral editing and
water suppression. NMR Biomed, 1998. 11(6): p. 266-72.
4. Hayes, D.J., et al., Brain gamma-aminobutyric acid: a neglected
role in impulsivity. Eur J Neurosci, 2014. 39(11): p. 1921-32.
5. Caprioli,
D., et al., Gamma aminobutyric acidergic
and neuronal structural markers in the nucleus accumbens core underlie
trait-like impulsive behavior. Biol Psychiatry, 2014. 75(2): p. 115-23.