0006-Ch11.pdf

Structural Plasticity of

Motoneuron Dendrites

Caused by Axotomy

P. Kenneth Rose, Victoria MacDermid, and

Monica Neuber-Hess

CONTENTS

11.1 Neuronal Polarity

11.2 Early Studies: Steps in the Wrong Direction?

11.2.1 Dendritic Tree Shrinkage: Then

11.2.2 Dendritic Tree Shrinkage: Now

11.2.3 Axotomy and Loss of Synaptic Integration

11.3 Steps in a Different Direction

11.3.1 Exceptions to the Rule: Dendritic Growth and Dendraxons

11.3.2 Supernumary Axons

11.4 Where Are We?

11.5 A Fresh Start: Steps in the Right Direction?

11.5.1 Dendritic Tree Expansion

11.5.2 Unusual Distal Dendrites

11.5.2.1 Light Microscopic Characteristics

11.5.2.2 Electron Microscopic Characteristics

11.5.2.3 Molecular Characteristics

11.6 Have We Arrived at Our Destination?

11.6.1 A New Perspective

11.6.2 Future Steps

11.7 Concluding Comments

References

11.1 NEURONAL POLARITY

Most neurons have two distinct compartments. As described almost 100 years ago

by Ramón Y Cajal, 1 the dendritic domain is composed of relatively short branches

that gradually taper and form acute angles between sibling branches. In contrast,

processes belonging to the axonal compartment travel for long distances with little

tapering and branches usually occur at right angles. Ultrastructurally, the most

conspicuous difference is visible at synapses where dendrites are most commonly

postsynaptic to presynaptic axonal specializations. 2 These structural differences are

matched by an equally distinctive set of molecular characteristics. 3 This morpholog-

ical and molecular polarity is related to a functional polarity where, with a few

notable exceptions, dendrites integrate synaptic signals and axons transmit action

potentials that trigger the release of neurotransmitters. 4 Thus, the concept of neuronal

polarity as a means of emphasizing the unique features of dendrites and axons,

whether at a structural, molecular, or functional level, has become a fundamental

principle of cellular neuroscience. The intricately orchestrated steps leading to the

formation of axons and dendrites in the developing nervous system 3,5 and the numer-

ous mechanisms that are involved in maintaining neuronal polarity 6,7 have reinforced

the importance attached to this integral property of neurons.

Implicit in most descriptions of the polarity of adult neurons is the assumption

that it is static. This assumption is questionable. The results of recent experiments

suggest that the structural and molecular polarity of adult neurons can be disrupted

following long-term axotomy. This plasticity has important consequences for under-

standing the capacity of the nervous system to recovery from neurotrauma. Much

of the evidence in support of a reorganization of neuronal polarity in the adult

nervous system is based on studies of permanently axotomized spinal motoneurons.

The goal of this chapter is to review and critically evaluate this evidence.

11.2 EARLY STUDIES: STEPS IN THE WRONG DIRECTION?

11.2.1 D ENDRITIC T REE S HRINKAGE : T HEN

The ﬁrst descriptions of the effect of axotomy on motoneuron structure provided no

evidence for alterations in their polarity. These studies, however, demonstrated that

dendritic structure of adult motoneurons was not ﬁxed.

In what would become a landmark paper, Sumner and Watson 8 reported that

axotomy leads to a reduction in the size of the dendritic tree of hypoglossal moto-

neurons. Dendrites of these motoneurons also retracted following intramuscular

injection of botulinum toxin, indicating that dendritic loss is partly a consequence

of a failure of neuromuscular transmission and not axonal damage alone. The

retraction was reversible upon reinnervation or upon regaining functional transmis-

sion several weeks after the botulinum toxin injection. The results of the axotomy

experiments were subsequently replicated in spinal motoneurons. 9 However, both of

these studies suffered from a serious methodological ﬂaw. At the time of these

studies, the full extent of the dendritic trees of motoneurons was not fully appreci-

ated. It was standard practice to describe motoneuron dendrites based on Golgi-

stained dendritic branches contained in a single histological section. Based on the

results of studies employing serial reconstructions of the dendritic trees of intracel-

lularly stained motoneurons, it is now apparent that this practice will exclude most

distal branches due to the complex three-dimensional distribution of the dendritic

trees of motoneurons. 10–14 As an example of the seriousness of this problem, Standler

m thick section. In

contrast, the average total length of intracellularly stained and reconstructed dendritic

trees of rat spinal motoneurons is 35.7 mm, 13 a 30-fold difference! Thus, the retrac-

tion of dendrites following axotomy, reported by Sumner and Watson, 8 may be an

artifact of the techniques used to measure dendritic tree size.

11.2.2D ENDRITIC T REE S HRINKAGE : N OW

In 1992, Brännström and colleagues 15,16 re-examined the issue of dendritic shrinkage

following axotomy of spinal motoneurons. This study avoided the short-comings of

the earlier investigations. All motoneurons were identiﬁed physiologically and mea-

surements were based on detailed reconstructions of intracellularly stained cells.

These measurements provided unequivocal evidence that a permanent axotomy of

12-weeks duration causes shrinkage of the dendritic trees of hindlimb motoneurons

in the adult cat. The shrinkage was evident in measurements of total dendritic length,

total surface area, and total number of dendritic branches ( Fig. 11.1 ). Cable theory

predicts that neuronal input resistance is inversely related to the size of the dendritic

trees. 17 Thus, the numerous electrophysiological reports of an increase in the input

resistance of motoneurons following axotomy 18–21 were consistent with the morpho-

logical results and provided a ﬁrm basis for the wide consensus that axotomy of

motoneurons leads to a decrease in the size of their dendritic trees.

11.2.3A XOTOMYAND L OSSOF S YNAPTIC I NTEGRATION

Dendritic retraction appears to be part of a larger, well-organized strategy to reduce

the integrative capacity of axotomized motoneurons. Other changes that led to the

same result include loss of synapses, 18–29 reduction in choline acetyltransferase, 30

and down regulation of neurotransmitter receptors. 31–34 Even the outputs of the

motoneuron are not immune to this transformation. Many of the axon collaterals

that form the basis for recurrent inhibition are lost following axotomy, 35 leaving the

motoneuron deprived of both inputs and outputs. Thus, the prevailing view of

axotomy-induced changes in motoneuron dendritic structure emphasizes degenera-

tive events and provides little basis for claims that one of the principal outcomes of

axotomy of spinal motoneurons is a change in their polarity. What has happened to

change this view?

11.3STEPS IN A DIFFERENT DIRECTION

11.3.1E XCEPTIONSTOTHE R ULE : D ENDRITIC G ROWTH

AND D ENDRAXONS

As described above, the remodeling of motoneuron dendritic structure following

axotomy is most consistent with a degenerative process. However, some morpho-

logical changes appear to contradict this conclusion. As described by Brännström

et al. 15 and shown in Fig. 11.2A , a small number (2 of 234) distal dendrites from

and Bernstein 9 reported an average total length of 1.14 mm for dendrites belonging

to rat spinal motoneurons that were captured on a single, 200-

FIGURE 11.1 Comparison of the change in dendritic tree size of hindlimb and neck moto-

neurons following permanent axotomy. Three indices of dendritic tree size are illustrated:

total dendritic length (the sum of the lengths of all dendritic segments for a single motoneu-

ron); total surface area (the sum of the surface areas of all dendritic segments for a single

motoneuron); and the total number of dendritic segments per motoneuron.The total number

of dendritic segments was not reported. This number was calculated based on their values of

the number of stem dendrites per motoneuron and the number of terminal dendrites per stem

dendrites, where the number of all dendritic segments per stem dendrite equals the number

of terminal dendrites + 1. (Data for axotomized neck motoneurons from Rose and Odlozinski,

J. Comp. Neurol., 390, 392, 1998.) As indicated by the direction of the arrows, axotomy

caused a decrease in the size of the dendritic trees of hindlimb motoneurons. In contrast, the

dendritic trees of neck motoneurons expanded. (Data for total dendritic length and surface

area for axotomized hindlimb motoneurons are based on the study of Brännström et al. 15

Comp. Neurol., 318, 439, 1992.)

axotomized motoneurons exhibited signs of expansion, instead of retraction. The

putative expansion appeared in two forms: a tangle of short preterminal and terminal

dendritic segments or a long, meandering, usually thick process that ended simply,

with no branches. The meandering trajectory is peculiar because this feature is a

hallmark of axons, not dendrites. Moreover, the swelling found at terminations of

last-order processes were, as described by Brännström et al., 15 typical of boutons on

axon collaterals. The similarity between these expanding dendrites and axons proved

to be only light-microscopically “deep”. Electron microscopic observations revealed

typical dendritic features, such as synaptic contacts, although contacts were rare on

one branch and dense collections of mitochondria were unusually frequent. 15

In the studies of Brännström et al., 15 the axon of the motoneuron was transected

close to muscle, a distance of approximately 15 to 20 cm from the soma. Cutting

motoneuron axons much closer to the soma led to a different result. In a brief report

that was subsequently followed by a more detailed description, Lindå, Risling, and

Cullheim 36,37 described the structural consequences of a parasaggital incision through

FIGURE 11.2 (A) Unusual distal dendrites of axotomized hindlimb motoneurons. The distal

dendrite in (i) gave rise to a complex arbor composed of numerous interwoven and sinuous

branches. An expanded view is shown in the inset. The distal dendrite in (ii) did not branch

and, unlike dendrites of intact motoneurons, followed a long and meandering path. (From

Brännström, T., Havton, L., and Kellereth, J.-O., J. Comp. Neurol., 316, 1, 1992. With

permission.) (B) Two examples of dendraxons. These axon-like processes (indicated by thick

lines) arose from dendrites of intraspinally axotomized hindlimb motoneurons and projected

toward the ventral roots. Only the portion of the dendritic tree that gave rise to the dendraxons

is shown. (From Lindå, H., Risling, M., and Cullheim, S., Brain Res., 358, 329, 1985. With

permission of Elsevier.) (C) Supernumary axons. The supernumary axon in (i) (double arrow)

projected dorsally and medially. Like axon collaterals of the original axon (single arrow), the

branches from the supernumary axon, formed numerous en passant and terminal boutons.

The supernumary axon in (ii) traveled rostrally in the lateral funiculus, for a distance of more

than 1 mm. The branches in the gray matter were thick and contorted. (From Havton, L. and

Kellerth, J.-O., Nature, 325, 711, 1987. With permission.) CC, central canal; L, lateral; V,

ventral; Cr, cranial; Ca, caudal.

m from their somata. Unlike dendrites of distally axotomized motoneurons,

where the vast majority of dendrites have a morphology typical of dendrites of

motoneurons with intact axons, many distal dendrites of intraspinally axotomized

motoneurons were aberrant. The most distinctive feature of these odd dendrites was

the lateral and ventral funiculi of L7. This injury transected motoneuron axons 400

to 1400

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: