SHORT COMMUNICATION Non-calyceal excitatory inputs mediate low ®delity synaptic transmission in rat auditory brainstem slices Martine Hamann, Brian Billups and Ian D. Forsythe Department of Cell Physiology and Pharmacology, University of Leicester, PO Box 138, Leicester LE1 9HN, UK Keywords: calyx of Held, medial nucleus of the trapezoid body, patch-clamp, presynaptic, glutamate Abstract Principal neurons of the medial nucleus of the trapezoid body (MNTB) receive a synaptic input from a single giant calyx terminal that generates a fast-rising, large excitatory postsynaptic current (EPSC), each of which are supra-threshold for postsynaptic action potential generation. Here, we present evidence that MNTB principal neurons receive multiple excitatory synaptic inputs generating slow-rising, small EPSCs that are also capable of triggering postsynaptic action potentials but are of non-calyceal origin. Both calyceal and non-calyceal EPSCs are mediated by a-amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA) and N-methyl-D-aspartate (NMDA) receptor activation; however, the NMDA receptor-mediated response is proportionally larger at the non-calyceal synapses. Non-calyceal synapses generate action potentials in MNTB principal neurons with a longer latency and a lower reliability than the large calyceal input. They constitute an alternative low ®delity synaptic input to the fast and secure relay transmission via the calyx of Held synapse. Introduction Themedialnucleusofthetrapezoidbody(MNTB)isanauditoryrelay nucleus that plays a role in sound localization. MNTB principal neurons are generally considered as a highly homogeneous group with a similar morphology and common functional properties. Each receives a somatic input from a single giant calyceal terminal (the calyxofHeld)generatingfast-risingandlargeexcitatorypostsynaptic currents (EPSCs; Barnes-Davies & Forsythe, 1995) that mediate a highlyreliable,phase-lockedsynaptictransmission(Taschenberger& von Gersdorff, 2000) suitable for high frequency auditory encoding. Previous reports have noted the presence of small inhibitory and excitatory synaptic responses (Banks & Smith, 1992; Forsythe & Barnes-Davies, 1993). Here we compare the excitatory non-calyceal and calyceal inputs after the onset of hearing and at physiological temperatures. We show that the non-calyceal input generates small slow rising EPSCs in MNTB principal neurons that can elicit post- synaptic action potentials (APs), although with less precision and lower ®delity than the calyceal input. Methods Preparation of brainstem slices Listerhoodedrats(postnatalday13±15)werekilledbydecapitationin accordance with UK animal experimentation regulations. Transverse brainstem slices (180±200 mm sections) were cut as previously described(Barnes-Davies&Forsythe,1995)andincubatedinarti®cial cerebrospinal ¯uid (ACSF) containing (in mM): NaCl, 125; KCl, 2.5; NaHCO 3 ,26;glucose,10;NaH 2 PO 4 ,1.25;CaCl 2 ,2;MgCl 2 ,1;myo- inositol, 3; sodium pyruvate, 2; ascorbic acid, 0.5, bubbled with 95% O 2 5% CO 2 (pH 7.4). Electrophysiology Cell-attachedandwhole-cellpatchclamprecordingsweremadeusing a MultiClamp 700 A ampli®er (Axon Instruments, USA). Recordings were made at a temperature of 35±36 8C in continuously perfused ACSF supplemented with 10 mM strychnine and 20 mM bicuculline to block inhibitory glycinergic and g-aminobutyric acid (GABA)ergic (mediated by GABA A receptors) synaptic inputs, respectively. Patch recording pipettes were constructed from thick-walled borosilicate glasscapillariesand®lledwithaninternalsolutioncontaining(inmM): KCl,130;Hepes,10;EGTA,5;ATP-Mg,2;GTP-Na 2 ,0.5;QX314,5; pH7.2 with KOH. EPSCs and APs were evoked using a bipolar platinum stimulation electrode positioned over the trapezoid body in the midline (Barnes-Davies & Forsythe, 1995). Stimulations (100- ms duration) were applied every 10 or 30s. EPSCs were recorded at a holding potential of 30mV. Series resistances (< 9MV) were compensated by 70±95%. Data analysis was conducted using Signal 2.05 software (Cambridge Electronic Design Ltd, UK). Sampling frequencies and ®lter settings were 10± 20kHz and 3.3±5kHz, respectively. Whole-cell membrane capaci- tance(18±35pF)wasdeterminedfromthedialsettingoftheampli®er aftercompensationofthetransientgeneratedbya10msvoltagestep. EPSCs were subsequently blocked by a cocktail of 50 mM AP5 and 10 mM NBQX[blocking N-methyl-D-aspartate(NMDA)and a-amino- 3-hydroxy-5-methyl-4-isoxazole propionate (AMPA) receptors, respectively]. In some experiments (Fig.2) EPSCs were recorded either in presence of the NMDA receptor antagonist (50 mM AP5) or in presence of the AMPA receptor antagonist (10 mM NBQX). EPSCs were then subsequently blocked by either 10 mM NBQX or 50 mM AP5,respectively,inordertodirectlycon®rmtheinvolvement European Journal of Neuroscience, Vol. 18, pp. 2899±2902, 2003 ß Federation of European Neuroscience Societies doi:10.1046/j.1460-9568.2003.03017.x Correspondence: Dr M. Hamann, as above. E-mail: mh86@le.ac.uk Received 6 June 2003, revised 29 August 2003, accepted 12 September 2003