Membrane Biophysics and Biochemistry NeuroReport 0959-4965 # Lippincott Williams & Wilkins Different Ca 2 source for slow AHP in completely adapting and repetitive ®ring pyramidal neurons J. C. Pineda, CA E. Galarraga 1 and R. C. Foehring 2 Centro de Investigaciones Regionales, Universidad Autonoma de Yucatan, Apdo. 9-37, Merida, Yuc. 97000; 1 Departamento de Bio®sica, Instituto de Fisiologia Celular, Universidad Nacional Autonoma de Mexico, 04510 Mexico; 2 Department of Anatomy and Neurobiology, University of Tennessee, Memphis, TN 38163, USA CA Corresponding Author INTRACELLULAR recordings in an in vitro neocortical slice preparation from immature rats were used to investigate the Ca 2 source for slow afterhyperpolariza- tion (sAHP) generation in pyramidal neurons that exhibit complete spike frequency adaptation (CA neu- rons). In pyramidal neurons that maintain repetitive ®ring for long periods of time (RF neurons), N-, P- and Q-type Ca 2 channels supply Ca 2 for sAHP genera- tion. In CA neurons, the sAHP was reduced by only 50% by the combination of antagonists for these Ca 2 channel types and L-type channels. Ryanodine and dantrolene, blockers of Ca 2 -induced Ca 2 release, reduced the sAHP by ,45% in CA neurons, but caused no reduction of the sAHP in RF neurons. Dantrolene application caused CA neurons to ®re throughout a 1 s suprathreshold current injection (as do RF neurons). NeuroReport 10:1951±1956 # 1999 Lippincott Williams & Wilkins. Key words: Ca 2 currents; Ca 2 -induced Ca 2 release; Dantrolene; Development; Neocortical slices; Ryanodine Introduction During the ®rst month of postnatal development, two types of pyramidal neuron can be distinguished on the basis of repetitive ®ring behavior [1,2] (Fig. 1A,B). Repetitive ®ring (RF) neurons exhibit rhyth- mic ®ring with variable spike-frequency adaptation (SFA) during long (e.g. 1 s) depolarizing current steps [1±4]. Completely adapting (CA) neurons exhibit strong SFA and cannot sustain ®ring throughout a 1 s depolarizing current step, regardless of the stimulus strength (often SFA is complete within 300 ms [1,2]. Repetitive spiking elicits Ca 2 -dependent after- hyperpolarizations (AHPs) in both RF and CA neurons [1±4]. The medium AHP (mAHP) does not differ between these cell types, but the slow AHP (sAHP) is larger and longer lasting in CA than in RF neurons [1±3]. Nonetheless, the sAHP in both neuron types is mainly due to Ca 2 -dependent K  conductances [1±4] (for general review see [5]). Recently, we reported that the combination of N-, P-, and Q-type Ca 2 current blockers blocked virtually all (median 94%) of the sAHP amplitude in RF pyramidal neurons [1]. During the course of those experiments, it appeared that the sAHP sensi- tivity to speci®c organic Ca 2 channel blockers was different in CA neurons. Therefore, we hypothesize that a different source of Ca 2 for sAHP generation is utilized by CA neurons. We found that only about half of the sAHP is reduced by the combina- tion of L-, N-, P- and Q-type current antagonists in CA neurons. The sAHP in CA neurons is sensitive (40±50% block) to ryanodine and dantrolene, sug- gesting that intracellular Ca 2 stores are an impor- tant source of Ca 2 for sAHP generation in CA neurons. The sAHP of RF neurons was not sensitive to ryanodine or dantrolene. Part of this work has been published in abstract form [6]. Materials and Methods Intracellular recordings were made from layer II±III sensorimotor neocortical neurons in an in vitro slice preparation from immature (postnatal day 7±23) Sprague±Dawley rats of both sexes. Our protocol conformed to the guidelines of the Animal Care and Use Committee, University of Tennessee, Memphis. Brie¯y, under anesthesia and on evidence of are¯ex- ia, the brains were quickly removed and submerged in a low sodium solution at 48C as described previously [1]. The low sodium solution included (in mM): 250 sucrose, 2.5 KCl, 1 NaH 2 PO 4 , 11 glucose, 4 MgSO 4 , 0.1 CaCl 2 and 15 HEPES (pH 7.3, 300 mOsm/l). Coronal neocortical slices (400 ìM) were cut with an oscillating tissue slicer and incubated in a 328C arti®cial cerebrospinal solution (ACSF) which contained (in mM): 125 NeuroReport 10, 1951±1956 (1999) Vol 10 No 9 23 June 1999 1951