Proceedings of The Physiological Society

University of Cambridge (2008) Proc Physiol Soc 11, PC111

Poster Communications

The use of viral vectors to examine projections from the periaqueductal grey to pontine noradrenergic neurones

L. Hickey1, S. Kasparov1, B. Liu1, E. Sher2, B. M. Lumb1, A. E. Pickering1

1. PhysPharm, Bristol University, Bristol, United Kingdom. 2. Eli Lilly, Surrey, United Kingdom.


The periaqueductal grey (PAG) is a key midbrain site involved in the modulation of nociception at the level of the spinal cord. However, projections from the PAG to the spinal cord are sparse and it is believed to control nociception, in part, by activating pontospinal noradrenergic (NA) neurones. To further understand the descending pathways from the PAG we have employed viral vectors to investigate the connections between the dorsolateral/lateral (DL/L-) PAG, and pontine NA neurones. In anaesthetised (Ketamine 60mg.kg-1/meditomidine 25µg.kg-1 i.p) male Wistar rats (n=5) injections of the adeno-associated viral vector AAV-CMV-eGFP (400nl) were made into the dorsolateral/lateral (DL/L) column of the PAG at sites at which prior injection of an excitatory amino acid (DL-homocysteic acid; 50Mm; 80nl) evoked ‘pressor’ responses. Animals were recovered for 8 days to allow time for anterograde transport of the viral vector to the pons. They were then terminally anaesthetised (sodium pentobarbital 70mg.kg-1 i.p.) and perfusion-fixed with 4% formalin. Brains were removed, post-fixed, and 40μm sections cut through the midbrain and pons. Sections were processed immunocytochemically to visualise terminals containing Green Fluorescent Protein (GFP) and to identify dopamine β-hydroxylase expressing NA neurones. Injection sites, terminal labelling and localisation of NA neurones were determined using conventional and confocal imaging. AAV-CMV-eGFP produced strong GFP labelling of PAG neurones (>92% NeuN +ve) and transfection extended within the PAG column in the rostrocaudal axis. GFP positive axons and terminals were seen in the pons with a predominantly ipsilateral distribution. Following injection into the DL/L-PAG the greatest number of terminals were seen in the pontine reticular area. Within the NA cell groups the strongest terminal labelling was seen in the rostral locus coeruleus (LC). There were also moderate projections to caudal LC and A7 regions with a low number of projections also noted in the A5 territory. Using both confocal microscopy and 3D imaging software (Velocity), many GFP labelled terminals in the LC and A7 territories were seen to closely appose both the somata and dendrites of NA neurones. In conclusion, using an adeno-associated viral vector, which has the advantage of being transported in the anterograde direction alone, we have been able to examine the connections of a functionally identified column of the PAG to regions of immunocytochemically identified NA neurones in the pons. The data support the view that neurones in the DL/L-PAG may exert their effects at the level of the spinal cord after engaging pontine NA centres, including LC.

Where applicable, experiments conform with Society ethical requirements