Pharmacological magnetic resonance imaging (phMRI) provides an approach to study effects of drug challenges on brain processes. Elucidating mechanisms of drug action helps us to better understand the workings of neurotransmitter systems, map brain function or facilitate drug development. phMRI is increasingly used in preclinical research employing rodent models; however, data interpretation and integration are complicated by the use of different experimental approaches between laboratories. In particular, the effects of different anaesthetic regimes upon neuronal and haemodynamic processes and baseline physiology could be problematic.This paper investigates how differences in phMRI research methodologies are manifested and considers associated implications, placing particular emphasis on choice of anaesthetic regimes.A systematic review of rodent phMRI studies was conducted. Factors such as those describing anaesthetic regimes (e.g. agent, dosage) and parameters relating to physiological maintenance (e.g. ventilatory gases) and MRI method were recorded.We identified 126 eligible studies and found that the volatile agents isoflurane (43.7 %) and halothane (33.3 %) were most commonly used for anaesthesia, but dosage and mixture of ventilatory gases varied substantially between laboratories. Relevant physiological parameters were usually recorded, although 32 % of studies did not provide cardiovascular measures.Anaesthesia and animal preparation can influence phMRI data profoundly. The variation of anaesthetic type, dosage regime and ventilatory gases makes consolidation of research findings (e.g. within a specific neurotransmitter system) difficult. Standardisation of a small(er) number of preclinical phMRI research methodologies and/or increased consideration of approaches that do not require anaesthesia is necessary to address these challenges.

The development of pharmacological magnetic resonance imaging (phMRI) has presented the opportunity for investigation of the neurophysiological effects of drugs in vivo. Psilocin, a hallucinogen metabolised from psilocybin, was recently reported to evoke brain region-specific, phMRI signal changes in humans. The present study investigated the effects of psilocin in a rat model using phMRI and then probed the relationship between neuronal and haemodynamic responses using a multimodal measurement preparation. Psilocin (2 mg/kg or 0.03 mg/kg i.v.) or vehicle was administered to rats (N=6/group) during either phMRI scanning or concurrent imaging of cortical blood flow and recording of local field potentials. Compared to vehicle controls psilocin (2 mg/kg) evoked phMRI signal increases in a number of regions including olfactory and limbic areas and elements of the visual system. PhMRI signal decreases were seen in other regions including somatosensory and motor cortices. Investigation of neurovascular coupling revealed that whilst neuronal responses (local field potentials) to sensory stimuli were decreased in amplitude by psilocin administration, concurrently measured haemodynamic responses (cerebral blood flow) were enhanced. The present findings show that psilocin evoked region-specific changes in phMRI signals in the rat, confirming recent human data. However, the results also suggest that the haemodynamic signal changes underlying phMRI responses reflect changes in both neuronal activity and neurovascular coupling. This highlights the importance of understanding the neurovascular effects of pharmacological manipulations for interpreting haemodynamic neuroimaging data.

The integrity of the white matter is critical in regulating efficient neuronal communication and maintaining cognitive function. Damage to brain white matter putatively contributes to age-related cognitive decline. There is a growing interest in animal models from which the mechanistic basis of white matter pathology in aging can be elucidated but to date there has been a lack of systematic behavior and pathology in the same mice. Anatomically widespread, diffuse white matter damage was induced, in 3 different cohorts of C57Bl/6J mice, by chronic hypoperfusion produced by bilateral carotid stenosis. A comprehensive assessment of spatial memory (spatial reference learning and memory; cohort 1) and serial spatial learning and memory (cohort 2) using the water maze, and spatial working memory (cohort 3) using the 8-arm radial arm maze, was conducted. In parallel, a systematic assessment of white matter components (myelin, axon, glia) was conducted using immunohistochemical markers (myelin-associated glycoprotein [MAG], degraded myelin basic protein [dMBP], anti-amyloid precursor protein [APP], anti-ionized calcium-binding adapter molecule [Iba-1]). Ischemic neuronal perikarya damage, assessed using histology (hematoxylin and eosin; H&E), was absent in all shams but was present in some hypoperfused mice (2/11 in cohort 1, 4/14 in cohort 2, and 17/24 in cohort 3). All animals with neuronal perikaryal damage were excluded from further study. Diffuse white matter damage occurred, throughout the brain, in all hypoperfused mice in each cohort and was essentially absent in sham-operated controls. There was a selective impairment in spatial working memory, with all other measures of spatial memory remaining intact, in hypoperfused mice with selective white matter damage. The results demonstrate that diffuse white matter pathology, in the absence of gray matter damage, induces a selective impairment of spatial working memory. This highlights the importance of assessing parallel pathology and behavior in the same mice.

Traumatic brain injury (TBI) is one of the most robust risk factors for the development of Alzheimer's Disease with epidemiological studies reporting a threefold increased risk (Fleminger et al, 2003). Furthermore, a neuropathological link has been demonstrated with extracellular β-amyloid plaques reported in approximately 30% of individuals who died following severe head injury (Graham et al, 1996; Ikonomovic et al, 2004). However, there is increasing evidence that intraneuronal amyloid is likely to be an early pathogenic event in AD. We sought to investigate whether a mild brain injury could result in changes in the levels of intraneuronal amyloid. Post mortem tissue from individuals with a history of head injury were identified from the Neuropathology Department at the University of Edinburgh. Paraffin embedded sections from the hippocampus of control individuals (n = 18 age matched controls) and individuals with TBI (survival time 0-24 hours n = 18, 1-28 days, n = 6, 1-32 years n = 13) underwent 4G8 immunohistochemistry to detect amyloid- β and AT8 immunostaining to detect neurofibrillary tangles. Sections underwent semi-quantitative assessment of intraneuronal amyloid and tau immunostaining using a scoring system of 0-3 in the dentate gyrus, CA4, CA3, CA1 and subiculum. Intraneuronal amyloid was detected in both control cases and in response to mild brain injury. There was a significant increase in intraneuronal amyloid in all fields of the hippocampus in the TBI cases compared with controls (p = 0.0027 -0.0135). Intense immunostaining (grade 3) was present in 0-15% of control cases and 45-55% of TBI cases. Intraneuronal amyloid increased with increasing survival time following TBI and was most extensive in individuals who survived longest after injury (p = 0.0035 - 0.0087 compared to controls). There was no evidence of increased neurofibrillary tangles in the head injury as compared to controls. This study indicates that mild brain injury is sufficient to induce alterations in amyloid and highlights the need to understand more about the mechanisms by which injury can lead to AD.

Mild traumatic brain injury (TBI) accounts for up to 80% of clinical TBI and can result in cognitive impairment and white matter damage that may develop and persist over several years. Clinically relevant models of mild TBI for investigation of neurobiological changes and the development of therapeutic strategies are poorly developed. In this study we investigated the temporal profile of axonal and somal injury that may contribute to cognitive impairments in a mouse model of mild TBI. Neuronal perikaryal damage (hematoxylin and eosin and Fluoro-Jade C), myelin integrity (myelin basic protein and myelin-associated glycoprotein), and axonal damage (amyloid precursor protein), were evaluated by immunohistochemistry at 4 h, 24 h, 72 h, 4 weeks, and 6 weeks after mild lateral fluid percussion brain injury (0.9 atm; righting time 167 +/- 15 sec). At 3 weeks post-injury spatial reference learning and memory were tested in the Morris water maze (MWM). Levels of damage to neuronal cell bodies were comparable in the brain-injured and sham groups. Myelin integrity was minimally altered following injury. Clear alterations in axonal damage were observed at various time points after injury. Axonal damage was localized to the cingulum at 4 h post-injury. At 4 and 6 weeks post-injury, axonal damage was evident in the external capsule, and was seen at 6 weeks in the dorsal thalamic nuclei. At 3 weeks post-injury, injured mice showed an impaired ability to learn the water maze task, suggesting injury-induced alterations in search strategy learning. The evolving localization of axonal damage points to ongoing degeneration after injury that is concomitant with a deficit in learning.

Cerebrovascular risk factors contribute to the development of white matter changes associated with cognitive impairment in the elderly. Moreover cerebrovascular dysfunction, due to chronic hypoperfusion, may contribute to the development of cognitive impairment and the pathogenesis of Alzheimer's disease. We hypothesise that hypoperfusion is a cause of chronic white matter dysfunction and accelerated cognitive decline and that this additionally precipitates Alzheimer pathology. We have developed an experimental mouse model of chronic cerebral hypoperfusion to test this and have firstly characterised the structural pathology and behavioural changes in wild-type mice. Adult C57Bl/6 male mice underwent common carotid artery stenosis using an external microcoil with a defined internal diameter (0.18mm) applied permanently to both common carotid arteries under anaesthesia. Sham-operated controls underwent the same surgery except the microcoils were not applied. At one month spatial reference memory was assessed using a Morris watermaze paradigm and working memory was assessed using an 8-arm radial maze test. At the end of behavioural testing, brain sections were assessed using immunohistochemistry with markers of myelin and axonal integrity, microglia and ischaemic neuronal damage. During training on the spatial reference memory task, hypoperfused mice learned the location of the hidden platform as quickly as controls and reached a comparable level of performance. Short-term and long-term spatial reference memory retention were unaffected after chronic hypoperfusion. However, in the working memory tasks, the hypoperfused mice made significantly more errors than the control mice and fewer novel arm entries in their first 8 arm entries as compared to shams. A greater extent of myelin damage was present after chronic hypoperfusion as compared to shams in white matter regions (predominantly corpus callosum, internal capsule, optic tract). This was associated with a subtle increase in the number of microglia. There was minimal damage to axons and neuronal perikarya. Chronic cerebral hypoperfusion impairs white matter integrity which, although not sufficient to compromise spatial reference memory, does impair working memory. This model may be appropriate to study the impact of changes in white matter integrity on cognitive ability and a basis to understand the impact of hypoperfusion on the development of Alzheimer's disease.

Widespread damage and dysfunction of axons and myelin tracts that comprise white matter in Alzheimer's disease (AD) has been demonstrated by numerous neuropathological, imaging and biochemical studies. The relationship of white matter changes to the development of pathological changes in AD brain that are characteristic of the disease (plaques, tangles and memory loss) and the underlying mechanisms are yet to be defined. Head injury is the major environmental risk factor for AD and can result in persistent cognitive deficits and damage to the white matter in humans. We developed a rodent model of head injury to examine the relationship between head injury, the resulting white matter damage and cognitive deficits, and Alzheimer's disease. The aim of the current study was to examine the relationship between axon and myelin damage with cognitive deficits in an in vivo mouse model of head injury. Mice (C57/BLJ6) underwent mild lateral fluid percussion injury (FPI) (n=10) or sham surgery (n=8). Animals were tested for cognitive deficits using the Morris water maze (MWM) at 4 and 8 weeks after injury. Animals were terminated by perfusion fixation 8 weeks following surgery and sections from 2 coronal anatomical levels in the brain (-1.94 and -2.18mm from bregma) were stained using APP immunohistochemistry to determine axonal damage, MAG immunohistochemistry to determine myelin damage, and H & E staining to determine neuronal cell body damage. Axon and myelin damage were assessed using a semi-quantitative scoring system. At 4 weeks, injured mice demonstrated spatial learning deficits as well as short term spatial memory deficits in the MWM. Behavioural testing at 8 weeks after injury revealed no differences between the groups. Axonal damage was detected at greater levels in FPI mice compared with controls, particularly in the thalamus and internal capsule ipsilateral to the injury site. In contrast, no differences in myelin and neuronal cell body damage were detected following FPI compared with controls. This study demonstrated that selective changes occur in the axons of white matter in response to mild head injury, which are associated with deficits in cognitive ability. The funding of the Alzheimer's Society is gratefully acknowledged.