European Journal of Neuroscience

Pamilo, Siina; Malinen, Sanna; Hotta, Jaakko; Seppä, Mika; Foxe, John

doi: 10.1111/ejn.13034pmid: 26226919

We present a simple but effective correlation‐based method (maxCorr) for extracting subject‐specific components from group‐fMRI data. The method finds signal components that correlate maximally with the data set of one subject and minimally with the data sets of the other subjects. We show that such subject‐specific components are often related to movement and physiological noise (e.g. cardiac cycle, respiration). We further demonstrate that removing the most subject‐specific components for each subject reduces the overall data variance and improves the statistical identification of true fMRI activations. We compare the performance of maxCorr with CompCor, a commonly used artifact‐finding method in fMRI analysis. We show that maxCorr is less likely than CompCor to remove actual stimulus‐related activity, especially when no information about the stimulus is available. MaxCorr operates without stimulus information and is therefore well suitable for analyses of fMRI experiments employing naturalistic stimuli, such as movies, where stimulus regressors are difficult to construct, and for brain decoding techniques benefiting from reduced subject‐specific variance in each subject's data.

Wild, A. R.; Bollands, M.; Morris, P. G.; Jones, S.; Bolam, Paul

doi: 10.1111/ejn.13075pmid: 26370007

N‐Methyl‐d‐aspartate glutamate receptors (NMDARs) contribute to neural development, plasticity and survival, but they are also linked with neurodegeneration. NMDARs at synapses are activated by coincident glutamate release and depolarization. NMDARs distal to synapses can sometimes be recruited by ‘spill‐over’ of glutamate during high‐frequency synaptic stimulation or when glutamate uptake is compromised, and this influences the shape of NMDAR‐mediated postsynaptic responses. In substantia nigra dopamine neurons, activation of NMDARs beyond the synapse during different frequencies of presynaptic stimulation has not been explored, even though excitatory afferents from the subthalamic nucleus show a range of firing frequencies, and these frequencies change in human and experimental Parkinson's disease. This study reports that high‐frequency stimulation (80 Hz/200 ms) evoked NMDAR‐excitatory postsynaptic currents (EPSCs) that were larger and longer lasting than those evoked by single stimuli at low frequency (0.1 Hz). MK‐801, which irreversibly blocked NMDAR‐EPSCs activated during 0.1‐Hz stimulation, left a proportion of NMDAR‐EPSCs that could be activated by 80‐Hz stimulation and that may represent activity of NMDARs distal to synapses. TBOA, which blocks glutamate transporters, significantly increased NMDAR‐EPSCs in response to 80‐Hz stimulation, particularly when metabotropic glutamate receptors (mGluRs) were also blocked, indicating that recruitment of NMDARs distal to synapses is regulated by glutamate transporters and mGluRs. These regulatory mechanisms may be essential in the substantia nigra for restricting glutamate diffusion from synaptic sites and keeping NMDAR‐EPSCs in dopamine neurons relatively small and fast. Failure of glutamate transporters may contribute to the declining health of dopamine neurons during pathological conditions.

Masi, Alessio; Narducci, Roberto; Resta, Francesco; Carbone, Carmen; Kobayashi, Kazuto; Mannaioni, Guido; Roeper, Jochen

doi: 10.1111/ejn.13066pmid: 26354486

The selective vulnerability of substantia nigra pars compacta (SNc) dopaminergic (DA) neurons is an enigmatic trait of Parkinson's disease (PD), especially if compared to the remarkable resistance of closely related DA neurons in the neighboring ventral tegmental area (VTA). Overall evidence indicates that specific electrophysiological, metabolic and molecular factors underlie SNc vulnerability, although many pieces of the puzzle are still missing. In this respect, we recently demonstrated that 1‐methyl‐4‐phenylpyridinium (MPP+), the active metabolite of the parkinsonizing toxin 1‐methyl‐4‐phenyl‐1,2,3,6‐tetrahydropyridine (MPTP), alters the electrophysiological properties of SNc DA neurons in vitro by inhibiting the hyperpolarization‐activated current (Ih). Here, we present an electrophysiological investigation of the functional role of Ih in the integration of synaptic inputs in identified SNc and VTA DA neurons, comparatively, in acute midbrain slices from TH‐GFP mice. We show that pharmacological suppression of Ih increases the amplitude and decay time of excitatory postsynaptic potentials, leading to temporal summation of multiple excitatory potentials at somatic level. Importantly, these effects are quantitatively more evident in SNc DA neurons. We conclude that Ih regulates the responsiveness to excitatory synaptic transmission in SNc and VTA DA neurons differentially. Finally, we present the hypothesis that Ih loss of function may be linked to PD trigger mechanisms, such as mitochondrial failure and ATP depletion, and act in concert with SNc‐specific synaptic connectivity to promote selective vulnerability.

doi: 10.1111/ejn.12989pmid: N/A

Cover Illustration: A horizontal cell of the mouse retina contacts almost all cone pedicles within its dendritic field. During patch‐clamp recording in a horizontal slice preparation, a horizontal cell was injected with Alexa 568 (red). The slice was fixed and counterstained with fluorescein‐labeled peanut‐agglutinin as a marker for cone pedicles (green). The image shows a three‐dimensional reconstruction of the soma and proximal dendritic tree of the horizontal cell placed on the cone pedicle distribution. Merging of both stainings reveals synaptic contacts between cones and horizontal cell dendrites. Eur. J. Neurosci., 42, 2615–2632.

Han, Su Young; Bouwer, Gregory T.; Seymour, Alexander J.; Korpal, Aaron K.; Schwenke, Daryl O.; Brown, Colin H.; Herman, James

doi: 10.1111/ejn.13062pmid: 26342194

Vasopressin secretion from the posterior pituitary gland is determined by action potential discharge of hypothalamic magnocellular neurosecretory cells. Vasopressin is a potent vasoconstrictor, but vasopressin levels are paradoxically elevated in some patients with established hypertension. To determine whether vasopressin neurons are excited in hypertension, extracellular single‐unit recordings of vasopressin neurons from urethane‐anaesthetized Cyp1a1‐Ren2 rats with inducible angiotensin‐dependent hypertension were made. The basal firing rate of vasopressin neurons was higher in hypertensive Cyp1a1‐Ren2 rats than in non‐hypertensive Cyp1a1‐Ren2 rats. The increase in firing rate was specific to vasopressin neurons because oxytocin neuron firing rate was unaffected by the induction of hypertension. Intravenous injection of the α1‐adrenoreceptor agonist, phenylephrine (2.5 μg/kg), transiently increased mean arterial blood pressure to cause a baroreflex‐induced inhibition of heart rate and vasopressin neuron firing rate (by 52 ± 9%) in non‐hypertensive rats. By contrast, intravenous phenylephrine did not inhibit vasopressin neurons in hypertensive rats, despite a similar increase in mean arterial blood pressure and inhibition of heart rate. Circulating angiotensin II can excite vasopressin neurons via activation of afferent inputs from the subfornical organ. However, the increase in vasopressin neuron firing rate and the loss of inhibition by intravenous phenylephrine were not blocked by intra‐subfornical organ infusion of the angiotensin AT1 receptor antagonist, losartan. It can be concluded that increased vasopressin neuron activity at the onset of hypertension is driven, at least in part, by reduced baroreflex inhibition of vasopressin neurons and that this might exacerbate the increase in blood pressure at the onset of hypertension.

Fujimoto, Hisataka; Ohgomori, Tomohiro; Abe, Kentaro; Uchimura, Kenji; Kadomatsu, Kenji; Jinno, Shozo; Kano, Masanobu

doi: 10.1111/ejn.13073pmid: 26369722

Keratan sulfate proteoglycans (KSPGs) and chondroitin sulfate proteoglycans (CSPGs) consist of a protein core with covalently attached glycosaminoglycan side chain. Although CSPGs are known to regulate the end of the critical period, the role of KSPGs in brain development remains unclear. Young male zebra finches memorise song templates during development. The brain regions that are responsible for song learning, known as song nuclei, are recognized as a suitable model for the study of brain development. To understand the potential role of KSPGs, here we examined the localization of KSs with different degrees of sulfation in the brain of developing male zebra finches. Exclusively in the song nuclei, an increase in expression of 5‐D‐4‐positive (5‐D‐4+) high‐sulfated KS started after hatching, and reached a plateau at the end of the sensory period, during which the young bird listens to and memorises the song of an adult tutor. By contrast, weak and ubiquitous expression of BCD‐4+ low‐sulfated KS remained unchanged until the end of the sensory period, and first increased in the song nuclei at the end of the sensorimotor period, during which the young bird produces plastic songs. Immunoblot analysis showed that phosphacan was a common core protein of 5‐D‐4+ KS and BCD‐4+ KS. Finally, we confirmed that the sulfotransferase responsible for the synthesis of high‐sulfated KS was exclusively localised in the song nuclei. Our observations suggest that time‐dependent localization of KSPGs with different sulfation patterns in the song nuclei may underlie song learning in developing male zebra finches.

Yoshikawa, Tomoko; Nakajima, Yoshihiro; Yamada, Yoshiko; Enoki, Ryosuke; Watanabe, Kazuto; Yamazaki, Maya; Sakimura, Kenji; Honma, Sato; Honma, Ken‐ichi; Silver, Rae

doi: 10.1111/ejn.13061pmid: 26342201

Feigenspan, Andreas; Babai, Norbert

doi: 10.1111/ejn.13016pmid: 26173960

As all visual information is represented in the spatio‐temporal dynamics of transmitter release from photoreceptors and the combined postsynaptic responses of second‐order neurons, appropriate synaptic transfer functions are fundamental for a meaningful perception of the visual world. The functional contribution of horizontal cells to gain control and organization of bipolar and ganglion cell receptive fields can only be evaluated with an in‐depth understanding of signal processing in horizontal cells. Therefore, a horizontal slice preparation of the mouse retina was established to record from horizontal cell bodies with their dendritic fields intact and receiving functional synaptic input from cone photoreceptors. Horizontal cell bodies showed spontaneous excitatory currents (spEPSCs) of monophasic and more complex multi‐peak waveforms. spEPSCs were induced by quantal release of glutamate from presynaptic cones with a unitary amplitude of 3 pA. Non‐stationary noise analysis revealed that spEPSCs with a monoexponential decay were mediated by 7–8 glutamate receptors with a single‐channel amplitude of 1.55 pA. Responses to photopic full‐field illumination were characterized by reduction of a tonic inward current or hyperpolarization, inhibition of spEPSCs, followed by a fast and transient inward current at light offset. The response to periodic dark/light transitions of different frequencies was dependent on the adaptational status of the cell with a limiting frequency of 10 Hz. Both on and off components of the light response were mediated by AMPA and kainate receptors. Detailed analysis of horizontal cell synaptic physiology is a prerequisite for understanding signal coding and processing at the photoreceptor ribbon synapse.

Seibt, Frederik; Schlichter, Rémy; Barrot, Michel

doi: 10.1111/ejn.13077pmid: 26370319

In the dorsal horn of the spinal cord (DH), noradrenaline (NA) is released by axons originating from the locus coeruleus and induces spinal analgesia, the mechanisms of which are poorly understood. Here, the effects of NA on synaptic transmission in the deep laminae (III–V) of the DH were characterized. It was shown that exogenously applied, as well as endogenously released, NA facilitated inhibitory (γ‐aminobutyric acid (GABA)ergic and glycinergic) synaptic transmission in laminae III–IV of the DH by activating α1‐, α2‐ and β‐adrenoceptors (ARs). In contrast, NA had no effect on excitatory (glutamatergic) synaptic transmission. Physical interruption of communications between deep and more superficial laminae (by a mechanical transection between laminae IV and V) totally blocked the effects of α2‐AR agonists and strongly reduced the effects of α1‐AR agonists on inhibitory synaptic transmission in laminae III–IV without directly impairing synaptic release of GABA or glycine from neurons. Short‐term pretreatment of intact spinal cord slices with the glial cell metabolism inhibitor fluorocitrate or pharmacological blockade of ionotropic glutamate and ATP receptors mimicked the consequences of a mechanical transection between laminae IV and V. Taken together, the current results indicate that the facilitation of inhibitory synaptic transmission in laminae III–IV of the DH by NA requires functional interlaminar connections between deep and more superficial laminae, and might strongly depend on glia to neuron interactions. These interlaminar connections and glia to neuron interactions could represent interesting targets for analgesic strategies.

Dicken, Matthew S.; Hughes, Alexander R.; Hentges, Shane T.; Kano, Masanobu

doi: 10.1111/ejn.13076pmid: 26370162

The strength of γ‐aminobutyric acid (GABA)‐mediated inhibitory synaptic input is a principle determinant of neuronal activity. However, because of differences in the number of GABA afferent inputs and the sites of synapses, it is difficult to directly assay for altered GABA transmission between specific cells. The present study tested the hypothesis that the level of mRNA for the GABA synthetic enzyme glutamate decarboxylase (GAD) can provide a reliable proxy for GABA release. This was tested in a mouse hypothalamic circuit important in the regulation of energy balance. Fluorescent in situ hybridization results show that the expression of Gad1 mRNA (encoding the GAD67 enzyme) was increased in hypothalamic neuropeptide Y/agouti‐related peptide (NPY/AgRP) neurons after an overnight fast, consistent with the ability of GABA from these neurons to stimulate food intake. Optogenetic studies confirmed that the observed increase in Gad1 mRNA correlated with an increase in the probability of GABA release from NPY/AgRP neurons onto downstream proopiomelanocortin neurons. Likewise, there was an increase in the readily releasable pool of GABA in NPY/AgRP neurons. Selective inhibition of GAD activity in NPY/AgRP neurons decreased GABA release, indicating that GAD67 activity, which is largely dictated by expression level, is a key determinant of GABA release. Altogether, it appears that Gad expression may be a reliable proxy of altered GABAergic transmission. Examining changes in Gad mRNA as a proxy for GABA release may be particularly helpful when the downstream targets are not known or when limited tools exist for detecting GABA release at a particular synapse.