Elsevier

Neurocomputing

Volumes 44–46, June 2002, Pages 13-18
Neurocomputing

Sensitivity to interstimulus interval due to calcium interactions in the Purkinje cell spines

https://doi.org/10.1016/S0925-2312(02)00361-2Get rights and content

Abstract

Pairing specific LTD (PSD) is produced by paired parallel fiber (PF) and climbing fiber (CF) stimulation and requires Ca2+ elevation. CF or PF activation cause Ca2+ increase through voltage dependent channels and IP3 induced Ca2+ release, respectively. We developed a model of Ca2+ dynamics in Purkinje cell spines to investigate why paired PF–CF activation is necessary for PSD. Simulations show a supralinear increase of the Ca2+ signal if the CF input occurs in a restricted time interval following the PF input. Ca2+ buffers significantly contribute to this phenomenon. This mechanism may be involved in the requirement of temporal specificity in classical conditioning.

Introduction

In classical conditioning, memory storage depends on the order as well as the timing of the conditioned and unconditioned stimuli. A commonly used experimental paradigm is conditioning of the rabbit nictitating membrane response [3]. A tone is used as the conditioned stimulus and an air puff to the eyes as the unconditioned stimulus. The cerebellum, which is crucial to memory storage, receives information about the tone from parallel fibres (PF) and receives information about the airpuff from climbing fibres (CF). Both PF and CF stimulation lead to Ca2+ increases in the Purkinje cell spines and dendrites, the former through inositol trisphosphate (IP3) induced Ca2+ release (IICR) while CF activation causes dendritic Ca2+ spikes.

As part of identifying the PF and CF interactions leading to memory formation, a PF and CF stimulation protocol in a slice preparation using 8PF pulses, 100Hz, followed by 3 CF pulses, 20Hz, has been developed [9]. The resulting PSD found is a Ca2+ as well as PKC dependent process [2]. The aim of the present study is to analyze the interaction between the PF and CF induced Ca2+ transients, using stimulation protocols similar to those in [9].

Section snippets

Methods

PF activation of mGluR receptors leads to activation of G protein, phospholipase C and production of IP3 [14], [13]. Calcium concentration in the spine is determined by IICR [6], buffers [5], and diffusion from the shaft:dCa_spinedt=β(c1(r1mo3h3+r2)(500−Ca_spine)−r3(Ca_spine)2/(K32+(Ca_spine)2))+diffusion,where:mo=(IP3/(IP3+d1))((Ca_spine)/(Ca_spine+d5)),Q2=d2(IP3+d1)/(IP3+d3),ho=Q2/(Q2+Ca_spine),τh=1/(a2(Q2+Ca_spine)),dh/dt=(ho−h)/τh.Purkinje cell data were used to adjust the parameters [5]. a

Results and discussion

Fig. 1 illustrates the amount of IP3 and Ca2+ produced from 1, 8 or 16PF pulses given at 100Hz. IP3, but not calcium, is linearly related to the number of PF pulses [11]. One PF leads to an insignificant increase in Ca2+ (see inset); 8PF pulses increase Ca2+ to 300nM. The amount of IP3 produced by 8pF pulses, 10μM [1], is the threshold for producing significant IICR in experiments [1], [5], [10]. Sixteen PF pulses produces a suprathreshold IP3 concentration; rapid calcium release causes buffer

Jeanette Hellgren Kotaleski received a M.S. in Medical Sciences from Umeå University in 1989, a M.S. in Engineering Physics from Kungliga Tekniska Högskolan in 1991 and a Ph.D. in Computer Science in 1998 on work focusing on modeling bursting mechanisms and coordination in motor systems. She is now a postdoc at the Krasnow Institute working on modeling biochemical pathways underlying classical conditioning.

References (14)

There are more references available in the full text version of this article.

Cited by (0)

Jeanette Hellgren Kotaleski received a M.S. in Medical Sciences from Umeå University in 1989, a M.S. in Engineering Physics from Kungliga Tekniska Högskolan in 1991 and a Ph.D. in Computer Science in 1998 on work focusing on modeling bursting mechanisms and coordination in motor systems. She is now a postdoc at the Krasnow Institute working on modeling biochemical pathways underlying classical conditioning.

Kim T. Blackwell received a B.S. in Biomedical Engineering from Boston University in 1981 and a V.M.D. in Veterinary Medicine in 1986, a M.S. in Systems Engineering in 1987, and a Ph.D. in Bioengineering in 1988 from the University of Pennsylvania. From 1988 to 1996, she was a senior research scientist at ERIM. Since 1996, she has been at George Mason University as an associative professor for Computational Sciences and Informatics, and affiliated with the Krasnow Institute. She is currently studying biological mechanisms of classical conditioning using computer modeling and biophysical techniques.

View full text