Reverse engineering the vestibular system [Elektronische Ressource] : intrinsic and synaptic contribution to signal processing in frog central vestibular neurons / vorgelegt von Christian Andreas Rössert

Christian Andreas Rössert

Sujets

• Gesundheit

Informations

• Publié par: ludwig-maximilians-universitat_munchen
• Publié le: 01 janvier 2010
• Nombre de lectures: 14
• Langue: English
• Poids de l'ouvrage: 7 Mo

Extrait

Reverse Engineering the Vestibular
System: Intrinsic and Synaptic Con-
tribution to Signal Processing in Frog
Central Vestibular Neurons
Christian Andreas R ossert
Munchen 2010Reverse Engineering the Vestibular
System: Intrinsic and Synaptic Con-
tribution to Signal Processing in Frog
Central Vestibular Neurons
Christian Andreas R ossert
Dissertation
an der Graduate School of Systemic Neurosciences
der Ludwig{Maximilians{Universit at
Munc hen
vorgelegt von
Christian Andreas R ossert
aus Munc hen
Munc hen, den 15.09.2010Erstgutachter: PD Dr.-Ing. Stefan Glasauer
Zweitgutachter: Prof. Dr. Hans Straka
Tag der mundlic hen Prufung: 25.11.2010Summary
Central vestibular neurons in the brainstem are responsible for the major computational
step that transforms head acceleration-related sensory signals into appropriate extraocular
motor commands for retinal image stabilization. These second-order vestibular neurons
(2 VN) receive convergent multimodal information from semicircular canal and otolith
organs as well as from the visual (optokinetic) system and neck proprioception. In frog,
2 VN can be separated into two subpopulations (tonic and phasic neurons) that show di-
verse dynamics in their subthreshold, impedance and spike discharge responses suggesting
processing of low-dynamic head motion components in tonic 2 VN and of high-dynamic,
nonlinear head motion components in phasic 2 VN. Furthermore, phasic but not tonic
2 VN receive an ipsilateral disynaptic feed-forward inhibition that concurs with their
highly dynamic membrane properties.
While information about intrinsic and synaptic properties of 2 VN is available as exper-
imental data, the functional implications of these results on vestibular signal processing
was still missing. Furthermore, information on the speci c roles of di erent ion channels
in shaping the intracellular response dynamics of 2 VN and their respective interaction
with emerging network properties was unknown. This doctoral thesis attempts to answer
these questions by combining electrophysiological measurements and quantitative compu-
tational modeling.
To reveal the di erent modes of signal processing in the two 2 VN subtypes, in a rst
study labyrinthine nerve branches to individual vestibular end organs have been stimu-
lated by trains of single electrical pulses imitating the spike activity of frog vestibular
nerve a erents during sinusoidal head rotation in vivo. With the help of this stimula-
tion paradigm, pronounced di erences in subthreshold dynamics and discharge behavior
in tonic and phasic 2 VN have been discovered. Additionally, quantitative subthreshold
Hodgkin-Huxley type models of 2 VN have been generated on the basis of available phys-
iological data and were supplemented by respective excitatory and inhibitory synaptic
inputs. With the help of this modeling approach it was possible to demonstrate that
feed-forward inhibition is the main factor in shaping the subthreshold dynamics of phasic
2 VN and the characteristics of the inhibitory interneurons could be determined.vi Summary
In the following study the combined subthreshold cellular and network model was ex-
tended by an integrate-and- re threshold to simulate spiking activity. Stimulating the
model with an activation pattern as recorded in vestibular nerve a erents during con-
stant angular head acceleration showed how intrinsic and synaptic components can shape
the ring behavior of phasic 2 VN. This study thus provided a rst insight into the coad-
aptation of intrinsic membrane properties and synaptic inhibition and generated testable
hypotheses for empiric experiments.
To obtain a better characterization of the membrane properties of frog 2 VN, in the third
study a frequency-domain analysis of 2 VN was conducted. However, in the course of
the experiments it became apparent that the errors induced by the sharp, high-resistant
electrodes used for the electrophysiological recordings in 2 VN could not be compensated
with the standard ampli er electrode compensation mechanisms. On this account, a
novel electrode compensation method called Piece-wise Non-linear Electrode Compensa-
tion (PNEC) has been developed that is capable of removing most electrode artifacts in
frequency-domain measurements.
In the last study, quantitative spiking Hodgkin-Huxley type models of phasic and tonic
2 VN based on frequency{domain data and interspike frequencies have been constructed.
To reproduce the di erential subthreshold- and spiking response behavior of phasic and
tonic 2 VN the two models di ered only by a single additional potassium channel that
was low-threshold voltage-dependent in phasic and spike-dependent in tonic 2 VN models.
As in the previous simulations, the spiking 2 VN models were extended by conductance-
based excitatory synapses and a feed-forward inhibition from tonic 2 VN interneurons
onto phasic 2 VN. The resulting quantitative models not only helped elucidating the role
of the di erent membrane parameters in shaping the intracellular response dynamics but
also revealed the relative contribution of intrinsic membrane and network properties to
synaptic signal processing in phasic 2 VN. Extension of the single cell models to a pop-
ulation model allowed making predictions on intrinsic and synaptic contributions during
more natural stimulation including asynchronous a erent ber input and synaptic noise.
In summary, the results of this thesis will help getting a better understanding of the
relative contributions of intrinsic and synaptic factors in generating dynamically appro-
priate motor commands for retinal image stabilization during locomotion. Furthermore
this thesis also has important implications for general electrophysiological measurements
using sharp, high-resistant electrodes and provides a promising new alternative electrode
compensation method.Contents
Summary v
1 Introduction 1
1.1 Semicircular canals, otholith organs and a erent nerve bers . . . . . . . . 1
1.2 Second-order vestibular neurons . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.1 Intrinsic membrane properties of second-order vestibular neurons in
amniotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.2 Intrinsic membrane properties of vestibular neurons in
frogs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.3 Inhibitory control of second-order vestibular neurons . . . . . . . . . . . . 11
1.4 Methodological aspect: Frequency-domain analysis . . . . . . . . . . . . . 13
1.5 Aim of this thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2 Cumulative Thesis 17
2.1 Di erential dynamic processing of a erent signals in frog tonic and phasic
second-order vestibular neurons . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2 Modeling of intrinsic and synaptic properties to reveal the cellular and
network contribution for vestibular signal processing . . . . . . . . . . . . . 21
2.3 Frequency-domain analysis of intrinsic neuronal properties using high-resistant
electrodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.4 Cellular and network contributions to vestibular signal processing: impact
of ion conductances, synaptic inhibition and noise . . . . . . . . . . . . . . 25
3 Discussion 27
3.1 Electrophysiology and quantitative modeling . . . . . . . . . . . . . . . . . 31
3.2 Intrinsic properties of second-order vestibular neurons of the frog . . . . . 33
3.3 Signal processing in the vestibular system of the frog . . . . . . . . . . . . 34
Bibliography 37
List of Publications 45
Acknowledgements 47
Enclosure 49
Di erential dynamic processing of a erent signals in frog tonic and phasic second-
order vestibular neurons . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Modeling of intrinsic and synaptic properties to reveal the cellular and network
contribution for vestibular signal processing . . . . . . . . . . . . . . . . . 67
Frequency-domain analysis of intrinsic neuronal properties using high-resistant
electrodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Cellular and network contributions to vestibular signal processing: impact of ion
conductances, synaptic inhibition and noise . . . . . . . . . . . . . . . . . . 871 Introduction
Vision, hearing, touch, taste and smell are usually the rst to be mentioned when some-
body is asked to list biological senses. The vestibular sense, which is the sense of motion
and orientation in space, is however often forgotten, yet it plays a vital role in our daily
life. The neglect of this sense is not surprising since it provides no distinct tangible sen-
sation and it cannot be shut o . We are thus not aware of this sense unless either sensory
con ict with other senses, mismatch or uncertainty in interpreting vestibular signals re-
sults in motion sickness or illusions. Other, even more severe incidents that illustrate
the importance of the vestibular sense are de cits in the vestibular system (e.g., due to
vestibular neuritis, lesions or hair cell loss) that result, amongst others, in disorientation,
imbalance and vertigo (Green and Angelaki, 2010).
The set of sensors for the vestibular sense, the vestibular organs, are located in the inner
ear on both sides of the head with a mirror-symmetrical spatial arrangement. The