Abstract: During the generation of rhythmic movements, most spinal neurons receive an oscillatory synaptic drive. The neuronal architecture underlying this drive is unknown, and the corresponding network size and sparseness have not yet been addressed. If the input originates from a small central pattern generator (CPG) with dense divergent connectivity, it will induce correlated input to all receiving neurons, while sparse convergent wiring will induce a weak correlation, if any. Here, we use pairwise recordings of spinal neurons to measure synaptic correlations and thus infer the wiring architecture qualitatively. A strong correlation on a slow timescale implies functional relatedness and a common source, which will also cause correlation on fast timescale due to shared synaptic connections. However, we consistently find marginal coupling between slow and fast correlations regardless of neuronal identity. This suggests either sparse convergent connectivity or a CPG network with recurrent inhibition that actively decorrelates common input.
The method paper on the turtle preparation that Rune Berg and I developed is now out. Title and abstract below.
Although it is known that the generation of movements is performed to a large extent in neuronal circuits located in the spinal cord, the involved mechanisms are still unclear. The turtle as a model system for investigating spinal motor activity has advantages, which far exceeds those of model systems using other animals. The high resistance to anoxia allows for investigation of the fully developed and adult spinal circuitry, as opposed to mammals, which are sensitive to anoxia and where using neonates are often required to remedy the problems. The turtle is mechanically stable and natural sensory inputs can induce multiple complex motor behaviors, without the need for application of neurochemicals. Here, we provide a detailed protocol of how to make the adult turtle preparation, also known as the integrated preparation for electrophysiological investigation. Here, the hind-limb scratch reflex can be induced by mechanical sensory activation, while recording single cells, and the network activity, via intracellular-, extracellular- and electroneurogram recordings. The preparation was developed for the studies by Petersen et al. (2014) and Petersen and Berg (2016), and other ongoing studies. [pdf]
Our paper is now out in eLife with an insight article by Mark Humphries. The story is featured on the eLife frontpage with cover illustration. Cover illustration shows a motoneuron loaded with biocytin (Cyan color. Nissl and ChAT stainings in magenta and yellow respectively). Histology image courtesy of Robertas Guzulaitis.
When spinal circuits generate rhythmic movements it is important that the neuronal activity remains within stable bounds to avoid saturation and to preserve responsiveness. Here, we simultaneously record from hundreds of neurons in lumbar spinal circuits of turtles and establish the neuronal fraction that operates within either a ‘mean-driven’ or a ‘fluctuation-driven’ regime. Fluctuation-driven neurons have a ‘supralinear’ input-output curve, which enhances sensitivity, whereas the mean-driven regime reduces sensitivity. We find a rich diversity of firing rates across the neuronal population as reflected in a lognormal distribution and demonstrate that half of the neurons spend at least 50% of the time in the ‘fluctuation-driven’ regime regardless of behavior. Because of the disparity in input-output properties for these two regimes, this fraction may reflect a fine trade-off between stability and sensitivity in order to maintain excitability across behaviors.
Today I submitted my PhD thesis with the title: The Neuronal Network Orchestration behind Motor Behaviors. It was supervised by Rune W. Berg.
In biological networks, millions of neurons organize themselves from microscopic noisy individuals to robust macroscopic entities. These entities are capable of producing higher functions like sensory processing, decision-making, and elaborate behavioral responses. Every aspect of these behaviors is the outcome of an advanced orchestration of the activity of populations of neurons. Through spiking activity, neurons are able to interact; yet we know little about how this interaction occurs in spinal networks. How is the activity distributed across the population? What is the composition of synaptic input that is received by the individual neurons and how is the synaptic input processed? This thesis focuses on aspects of these questions for spinal networks involved in the generation of stereotypical motor behaviors.
The thesis consists of two studies. In the first study, I investigated the synaptic input to motoneurons during rhythmic motor behaviors, and specifically the hypothesis that motoneurons receive concurrent excitatory and inhibitory (E/I) inputs. Berg et al. (2007) presented the concurrent hypothesis, which goes against the classical feed forward reciprocal model for spinal motor networks that has gained widespread acceptance. We developed an adult turtle preparation where the spinal motor network was intact, which also allowed us to perform intracellular recordings from motoneurons during rhythmic motor activity. We estimated the synaptic excitatory and inhibitory conductances by two individual methods. We found that the synaptic input to motoneurons primarily consisted of concurrent exci- tation and inhibition. We quantified the similarities found in intracellular recordings from the intact preparation with a transected turtle preparation, which was missing part of the spinal network. The transected preparation was used in Berg et al. (2007). We found no significant difference between the intracellular recordings from motoneurons performed in the two preparations. A spinal transection had marginal effects on the network integrity and the synaptic fluctuations were unaltered. We imple- mented a balanced network model that is widely applicable to data from cortex, to explain our results from the spinal cord. We found that this network was highly resilient to structural division. Both the experimental findings and the network model support that motoneurons receive concurrent synaptic E/I during behaviors from a premotor network with recurrent connections, which is operating in the irregular regime.
Our experimental findings are in agreement with studies from the cortex and the balanced model. It is therefore relevant to study the population activity in the spinal cord for traits from cortex studies. In the second study, we looked into the distribution of firing rates during different motor programs and the mechanisms that give rise to the distribution. We implanted high-density silicon probes in the spinal cord and recorded parallel single unit activity while inducing different scratch behaviors. We find that neuronal populations in the spinal cord have highly skewed distributions of firing rates. The majority of the neurons was spiking at low firing rates, while a minority had an activity level that was much higher. The distribution is lognormal-like and robust across trials, and it remains skewed in different behaviors. Our findings support that the neuronal activity, which is involved in motor behavior, is governed by synaptic fluctuations and as a result thereof is irregular. Similar lognormal- like distributions of firing rates have also been observed in other brain areas (Buzsáki and Mizuseki, 2014). Roxin et al. (2011) detailed the firing rate distribution in networks in the balanced regime, and found it to be similar to a lognormal distribution and describing the data from the population studies very well.
Our experimental observations and analysis are in agreement with the balanced model proposed by van Vreeswijk and Sompolinsky (1996) and Roxin et al. (2011). The studied spinal neuronal network is predominantly operating in the irregular subthreshold regime, yet neurons are pushed towards the regular superthreshold regime with increased synaptic input.