Combining mouse genomics and functional magnetic resonance imaging (fMRI) provides a promising tool to unravel the molecular mechanisms of somatosensation and pain. Recent studies suggest a confounding influence of mean arterial blood pressure (MABP) on the blood oxygenation level-dependent (BOLD) signal for somatosensory stimulation paradigms in mice^1,2. In rats, it was shown that MABP alterations induced by transient stimuli are prone to evoke specific BOLD patterns, which can mimic brain activation when autoregulation mechanisms are impaired – for instance by the application of anesthesia^3-6. Here we provide for the first time data, which rely on the monitoring of MABP for mild noxious (heat) stimulation in mice. Investigating the influence of MABP on the BOLD signal is crucial to interpret the suitability of somatosensory stimulation tasks in mouse fMRI.

Animal preparation: Six male C57BL/6N mice (weight 25-28 g) were studied using thermostimulation (46°C) under 1% isoflurane anesthesia. The animals were intubated, artificially ventilated and paralyzed using the neuromuscular blocking agent pancuronium bromide (1mg/kg) in order to avoid motion artifacts. A customized mouse cradle was used, which permits uniform, non-contact warming of the mouse⁷. A whole-body temperature of 36.3±0.5°C was maintained. Thermal stimulation was carried out for the plantar hindpaw using a feedback-controlled, water-cooled Peltier element⁷. Stimulation paradigm: 35.5°C baseline, 4°C/s, 46°C peak, 20s, error <0.2°C.

MR Imaging. High- resolution sagittal T2 weighted imaging were used to position 11 axial slices used for T₂*-weighted fMRI (GE-EPI, TR/TE/FA = 2500ms/11.0ms/80°, FOV/mtx/res = 24x12x5mm / 90x60x11 / 267x200x500μm), TA = 12min. All images were acquired on a 9.4T Bruker Biospec (Ettlingen, Germany) using a transceive cryogenic quadrature RF surface coil (Bruker, Ettlingen, Germany).

Data analysis. FMRI data were motion corrected, smoothed, registered to a mouse brain atlas and statistically analyzed at a group level (mixed effects (FLAME)) using FSL (http://www.fmrib.ox.ac.uk/fsl).

Monitoring of MABP and heart rate. Animal preparation was conducted as described above. The left femoral artery of three mice was cannulated to allow for continuous monitoring of MABP. Mice were positioned in the customized mouse cradle to study MABP and heart rate in response to subsequent heat stimuli under conditions similar to the fMRI experiments.

We found bilateral activations in cortical and subcortical areas in all mice (Fig. 1) when applying short, transient heat stimuli to the plantar paw. Z-statistic maps show only positive significance patterns, which span multiple brain areas when statistical thresholds are reduced to uncorrected inference (Fig. 1, bottom panel). Significant increases in MABP were observed in response to heat stimuli with a maximum alteration of approx. 20 mmHg (Fig. 2, left panel). Alterations in heart rate were also observed (Fig. 2, right panel).Spatial BOLD patterns for heat stimulation are located in brain areas well known for the processing of somatosensation and nociception^1,2,8 (Fig. 1). Bilateral cortical activations, exclusively positive BOLD patterns, and broad unspecific significance patterns at lower statistical thresholds are in line with previous mouse fMRI studies using electrostimulation¹. These observations could suggest an influence of MABP on the BOLD signal. Indeed, MABP alterations show similarities with the temporal course of the BOLD signal (Fig. 1, Fig. 2). However, it is not unexpected that salient events evoke increases in MABP^4,6. The efficiency of autoregulatory mechanisms to buffer influx of oxygenated blood into the brain strongly depends on the range and transition time of systemic MABP changes^3-6. An operational window of MABP alterations, which ensures the BOLD signal unaffected by systemic MABP remains to be determined for somatosensory and noxious stimuli in mice under isoflurane anesthesia.To shed a light on this issue, we currently work on a study to monitor BOLD and MABP simultaneously. We intent to further conduct cortical cell recordings to gain a direct evidence for brain activity: The wide-spanned BOLD patterns, which we observed could be alternatively explained by unspecific neuronal arousal evoked by an alerting reaction to a salient event^8,9. This is not unlikely since the applied dose of anesthesia was low in this and former studies, which used somatosensory stimulation in mice^1,2. The effective influence of MABP on hemodynamic readouts has to be further detailed to provide mouse fMRI as a reliable tool to study the molecular mechanisms of somatosensation and pain.