Putative propriospinal modulation of premotor and motor cortical output during grasping

The primary motor cortex (M1) and ventral premotor cortex (PMv) play a major role in the control of grasping. Anatomical studies have revealed that these regions project to the spinal cord, directly and indirectly, and likely interact with the propriospinal network (PN). The PN is a pre-motoneuronal network located at mid-cervical levels (C3-C4), which transmits and alters descending cortical commands for targeted reaching and grasping. How the PN interacts with motor output from M1 and PMv during different grasps in humans is not well understood. The PN can be studied indirectly by conditioning motor evoked potentials (MEPs), elicited by transcranial magnetic stimulation (TMS), and H-reflexes, elicited by peripheral nerve stimulation (PNS), with sub-threshold PNS. In experiment #1, sub-threshold PNS was applied to the ulnar nerve at the wrist to condition Flexor Carpi Radialis (FCR) MEPs, elicited by M1 TMS, during an isolated FCR contraction (iFCR) or FCR contraction with precision grip (PG) or whole hand grasp (WHG). Central and peripheral conduction times were used to time the arrival of descending and ascending volleys at the spinal cord at 5 different inter-stimulation intervals (ISIs), namely 0, -3, -4, -5 and -6 ms. Negative ISIs indicate that PNS is delivered prior to TMS, allowing time for PNS volleys to travel to higher cervical segments (e.g. C3-C4), and 0 ms indicating that PNS and TMS-evoked volleys converge monosynaptically at spinal motoneurons (C6-C8). In experiment #2, TMS was applied over PMv to condition PN modulated H-reflexes, elicited by PNS to the median nerve applied at the elbow, during iFCR, PG and WHG. Here, TMS and sub-threshold PNS ulnar nerve volleys were timed to arrive at PN levels (C3-C4) at 5 different ISIs, namely 0, 2, 4, 6 and 8 ms. H-reflexes were either elicited alone (baseline), or conditioned by ulnar nerve PNS (ISI: 4ms), or by ulnar PNS and PMv TMS. In experiment #1, we found a significant interaction between ISI and grasp. Specifically, at 0 ms ISI, MEPs were significantly larger during PG than WHG and, at -4 ms ISI, MEPs were significantly larger during WHG compared to PG. In Experiment #2, we found that PMv TMS differentially modulated H-reflexes during iFCR and PG, but only for late ISIs. This contrasts with our previous findings that, at rest, PMv interacted with PN at early ISIs. Our results suggest that while monosynaptic corticospinal pathways contribute to precisions grip, motor cortical output during whole hand grasp can be modulated by the PN. Interestingly, PMv appears to modulate the PN directly or indirectly (e.g. via M1) depending on whether these interactions are tested at rest or during contraction, respectively