Neural correlates of behavioral responses to pain in nerve growth factor beta mutation carriers

Abstract

Introduction: The selection and modification of behavioral responses to noxious events are crucial components of pain perception (Perini et al. 2013). Forecasting whether a nociceptive input mediate a real threat for tissue damage is finely taken care by central mechanisms (Morrison et al. 2013); but what happens when the peripheral signal is altered? We addressed this question by investigating a population of carriers of a rare mutation affecting the nerve growth factor beta (NGFB) gene (Einarsdottir et al. 2004) and comparing them to age matched controls. The mutation causes a moderate loss of thinly myelinated A∂ fibers and a severe loss of unmyelinated C afferents, leaving large myelinated Aß fibers intact (Minde et al., 2006). The carriers show different levels of impaired pain sensitivity although their pain thresholds don't differ from controls. We hypothesized that these mutation carriers may perceive pain sensations but can less efficiently translate these sensations into action, perhaps resulting in failure to make appropriate behavioral adjustments in potentially tissue-damaging situations. Methods: We collected functional and structural (T1 and diffusion weighted) scans and we carried a quantification of the fiber loss in the carriers using corneal confocal microscopy (CCM), a non-invasive method that reliably detects and quantifies C-fibre loss in neuropathy (Quattrini et al., 2007; Tavakoli et al., 2012). During fMRI, participants (N=12 carriers, N=12 age matched controls) received painful and non-painful thermal stimulations and were instructed to press a button at a visual cue onset only when perceiving painful (2 out of 4 runs) or non-painful (2 out of 4 runs) stimulations. This allowed modeling of voluntary motor responses and painful stimulation separately. Using signal detection theory we calculated their sensitivity to the behavioral task (d') that is their ability to distinguish painful from non-painful stimulations. After the fMRI session we looked at their "urge to move" by creating a behavioral design in which they could continuously rate their urgency to move away from a painful stimulation. Results: Carriers showed lower sensitivity to the task (d') and rated their urgency to move significantly lower than controls. For the main effect of pain regardless of motor response, the control group showed a robust bilateral activation of anterior insula (AI), whereas the carriers showed a remarkable reduced activation here. Right AI and right inferior frontal gyrus were positively correlated to the carriers' sensitivity to the task (d') suggesting an involvement of those areas in discriminating painful versus non painful stimulation (Fig1). The mid cingulate cortex (MCC) (corresponding to the caudal part of the cingulate motor zones (CMZ)) was activated for both painful stimulation and motor response (button press) and survived the conjunction analysis between pain and motor regressors confirming previous finding on the action-related role of MCC during pain (Perini at al. 2013). In addition MCC was positively correlated to the urgency to react to painful stimulation during the response interval. Using tract based spatial statistics (TBSS) we found a positive correlation in the white matter surrounding the MCC to the density of nociceptive fibers in the periphery, suggesting that the lower the density of fibers in the periphery, the lower the FA in this area (Fig2).