Self-Organizing Body Maps in the Spinal Cord

Sammanfattning: During development primary afferents grow into and establish neuronal connections in the spinal cord, conveying information of the anatomical and mechanical properties of the body. This body representation forms the basis for how we perceive sensory information and control our movements. Subsequently, the strength of the sensory input to spinal sensorimotor circuits is functionally adjusted by an experience dependent mechanism termed somatosensory imprinting. In this learning, tactile feedback on spontaneous movements play a key role in setting the strength of nociceptive input. However, the cellular mechanisms underlying this learning and the resulting action-based body representation has hitherto not been clear. Of particular interest was to clarify if structural plasticity is involved in somatosensory imprinting. In sharp contrast to the prevailing view of an innate body representation, a transitory floating organization was found in the neonate, subsequently undergoing profound reorganization through gradual establishment of new connections and elimination of erroneous axonal arbours. In this process, afferent fibres carrying different modalities seem to follow each other into the dorsal horn and terminate in close proximity during the first postnatal weeks. The final termination pattern of thin fibres depends on activity in tactile fibres, as disturbing the tactile input profoundly affects not only tactile terminations but also those of thin fibres, indicating cross-modality interactions. Importantly, these structural changes occur in parallel to the functional adaptation by somatosensory imprinting. Blocking NMDA receptors during this time period freeze the juvenile termination pattern and abolish functional reflex tuning, demonstrating for the first time that somatosensory imprinting is NMDA dependent. A variance analysis provided additional support for the notion that differences in strength between adult afferent connections to spinal sensorimotor circuits are due to differences in the number of connections rather than heterogeneity in gain between individual connections. NMDA receptors were also shown to set the gain in these circuits as well as in ascending nociceptive pathways to the primary somatosensory cortex. In conclusion, this thesis reveals that the spinal cord undergoes substantial activity dependent structural changes during development and suggests that these changes play a key role in somatosensory imprinting. We also show that NMDA receptors are important for gain regulation in the adult, both in spinal sensorimotor circuits as well as in ascending sensory processing to the somatosensory cortex. Thus, far from being inborn and stereotypic, the dorsal horn of the spinal cord should be regarded as a highly adaptive body-brain interface undergoing substantial experience dependent structural adaptation to the body anatomy and movement patterns.