Multidimensional Scaling of Drug Effects in CorticoStriatoThalamic Circuits

1
Dopaminergic drugs modulate specifically
fronto-striatal loadresponse in a paired
associate learning paradigm
DE Welchew(1), L Clark(2,3), R Cools(2,3), J
Suckling(1), ET Bullmore (1,3) and TW
Robbins(2,3)
(1)Brain Mapping Unit, Department of Psychiatry
and (2) Department of Experimental Psychology
University of Cambridge, UK (3)MRC Cambridge
Clinical and Behavioural Neurosciences Centre, UK
E-mail dew22_at_cam.ac.uk
RESULTS
INTRODUCTION

• The role of dopaminergic modulation of forebrain
  cortico-striato-thalamic circuits by mesocortical
  projections from the ventral tegmental area and
  substantia nigra (SN) has been studied in depth
  in both animal and human models. Anatomical and
  functional evidence exists for a circuit linking
  dorsolateral prefrontal cortex (PFC) and dorsal
  striatum involved in working memory.
• Sulpiride is a dopamine D2 receptor antagonist
  widely used in the treatment of schizophrenia,
  whilst methlyphenidate is an indirect monoamine
  agonist used in the treatment of ADHD. These
  drugs might therefore be expected to have
  opposing modulatory effects on dopaminergic
  circuits, and have recently been shown to alter
  practice and difficulty effects in an
  object-location learning task in elderly
  volunteers1. Here, a variant of this paradigm has
  been used to investigate the effects of
  dopaminergic modulation on the response to
  learning and task difficulty in young volunteers.

• Above left Clusters significantly activated by
  load, and
• Above right Clusters significantly activated by
  learning in the placebo condition.

METHODS
Imaging and Subjects

• Gradient echo echo-planar imaging at 3 Tesla
  (repetition time 1.1s) was used to study 12 (9
  male mean age 26.1 years, SD 3.8 years)
  performing a paired-associate learning task.
• The subjects were treated with methylphenidate
  (20mg), sulpiride (400mg) or placebo on each of
  three separate occasions. Treatment order was
  randomised across subjects.
• We used a periodic BACA design with eight
  repetitions of four 30 second blocks.
• A crosshair fixation. B identification of the
  location of a probe item in a previously
  presented array of two items. C as B with an
  array of four items.
• Probe locations remained constant throughout the
  experiment, the B and C conditions each consisted
  of 6 trials.

• Above left Dopaminergically modulated load
  response in the right caudate and precuneus.
• Above right Boxplots of individual subject load
  responses under each drug condition (all means
  are significantly different at plt0.05 corrected).
• Drug effects on learning were spread throughout
  parietal and motor cortex, but were not localised
  to striatal circuits.

CONCLUSIONS

• These results show that adaptivity of
  cortico-striato-thalamic systems to cognitive
  load is sensitive to modulation by dopaminergic
  drugs, whereas the role of these circuits in
  learning is unaffected.
• This supports the theory that dopaminergic inputs
  to subcortical nuclei, as well as the prefrontal
  cortex, are important in configuring
  cortico-striatal circuits to adapt to changes in
  load.
• The opposing effects on right striatal load
  response of sulpiride and methylphenidate are
  analogous to their opposing effects on
  dopaminergic transmission synaptically.

Analysis

• Functional brain activation mapping2 was used to
  perform a regression analysis modelling three
  mutually orthogonal aspects of brain activation
  at each voxel
• (1) between task and rest conditions
  (activation)
• (2) between the two difficulty levels (load
  response) and
• (3) across epochs in order of their presentation
  (familiarity, or learning).
• Statistic maps representing each of these effects
  were registered by affine transformation to a
  template image.
• Between-groups analysis was performed by fitting
  an ANOVA model including a single three-level
  factor for treatment at each intracerebral voxel
  to each subject, with treatment effects
  identified by a cluster-level permutation test
  (plt0.01).
• Post-hoc t-tests were used to examine the role of
  each drug in the two clusters identified for the
  load contrast.

REFERENCES

• 1Bullmore, E.T. et al. (2003). Cerebral Cortex,
  13144-154
• 2Brammer, M.J. et al. (1997). MRI, 15763-770

• This poster is available electronically
  athttp//www-staff.psychiatry.cam.ac.uk/dew22/h
  bm2003-welchew.ppt

DEW was supported by an MRC PhD studentship.