Final Report Summary - PAIN MODULATION (Neurobiological mechanisms of endogenous pain modulation) Executive Summary:Chronic pain is not only a serious burden for the individual suffering from it, but also one of the largest medical health problems worldwide – in Europe for example, it affects around 20% of adults and imposes costs of more than 200 billion Euros per year. One of the factors that have been suggested to play a role in chronic forms of pain is a dysfunction of the individual’s capability to modulate pain. Such endogenous control of painful sensations is realized by a brain system that originates in cortical areas and – via relays in the brainstem – targets neurons in the spinal cord. The spinal cord is the first station in the central nervous system where painful signals are processed and a modulation of these signals at such an early stage will produce a powerful effect on the signals travelling upwards to the brain and thus on the subsequent experience of pain. I first made use of recent innovations in data acquisition and analysis techniques to non-invasively investigate responses in the human spinal cord with the technique of functional magnetic resonance imaging (fMRI). Although it is highly difficult to investigate the spinal cord with fMRI (e.g. due to its small diameter and its proximity to the lungs), I was able to accurately pinpoint the location of responses to painful stimuli and their specificity in a first set of studies: 1) responses to painful stimulation are localized in the back of the spinal cord, in contrast to motor responses which are localized in the front of the spinal cord, 2) responses to painful stimulation of one side of the body are localized in the same side of the spinal cord, and 3) responses progress from higher up to lower down in the spinal cord when applying painful stimuli to the fingers from the thumb to the little finger. All of this is in accordance with what it known from invasive animal studies and human neurosurgery studies by other groups, but – crucially – I was able to non-invasively demonstrate these fundamental aspects of the neuroanatomy of the human spinal cord. Having established these basic neuroanatomical features, I next turned to investigate the influence of pain-modulatory systems on spinal cord responses. Here, I made use of the fact that somatosensory nerve-fibres from the peripheral nervous system are the only direct sensory input into the dorsal horn of the spinal cord, meaning that spinal cord neurons should for example not respond to visual stimuli. If however, the spinal cord responded to visual stimuli, this would mean that there is some descending modulatory input from the brain that relays visual signals to the spine. I hypothesized that such responses would only occur if the visual stimuli had some relation to pain, i.e. if a volunteer saw a certain visual signal that predicted pain. To this end, I used fMRI of the human spinal cord in combination with an associative learning task, in which visual stimuli were either predictive or non-predictive of painful stimulation and volunteers would thus form an expectation of impending pain or not. In several behavioural measures I observed that our volunteers learned the relationship between visual and painful stimuli and thus expected pain to follow one of these stimuli. Interestingly, I saw that the stronger a participant expected to receive pain, the weaker were the responses to pain in the spinal cord. Most importantly, I observed that the spinal cord also responded to visual signals if these were predictive of the upcoming pain. This relationship was furthermore dynamic, in that when volunteers were not expecting to receive pain anymore (i.e. when visual stimuli were no longer followed by pain), the spinal cord response to the visual stimulus decayed. These results demonstrate that there is a top-down pathway from the brain to the spinal cord that modulates the activity of spinal cord neurons in accordance with the expectations one holds with regard to upcoming pain. In another study aiming at supraspinal structures, I used a pharmacological manipulation of the opioidergic system (which is a crucial player in pain modulation) and could show that blocking opioidergic function with the receptor antagonist naloxone actually hinders the decay of neural responses to the visual stimulus – i.e. volunteers’ brains behave as if they still expect the pain to occur. Currently, I am investigating cortical responses underlying pain modulation with similar experiments using the tool of magnetoencephalography (MEG). This allows us to gain more detailed insights into the temporal profile of modulatory processes, as MEG possesses a time-resolution in the order of milliseconds as opposed to several seconds in fMRI. Here, I am especially interested in very fast responses in the gamma-frequency range, which have been suggested to be especially relevant for the subjective aspects of pain. Overall, our results delineate some of the pain modulatory capabilities our central nervous system possesses in a healthy state. The obtained results will provide guidance for future studies that seek to investigate these mechanisms in patients with chronic pain where they are assumed to be dysfunctional.