Effects of Dorsal Root Transection on Morphology and Chemical Composition of Degenerating Nerve...

Experimental Neurology 164, 236–245 (2000)doi:10.1006/exnr.2000.7429, available online at http://www.idealibrary.com on

Effects of Dorsal Root Transection on Morphology and ChemicalComposition of Degenerating Nerve Fibers and Reactive

Astrocytes in the Dorsal FuniculusRomuald Wroblewski,* Godfried M. Roomans,* and Elena N. Kozlova†

*Department of Medical Cell Biology and †Department of Neuroscience, Biomedical Center,University of Uppsala, 75123 Uppsala, Sweden

Received October 16, 1999; accepted January 18, 2000

The morphology and chemical (elemental) composi-tion of the dorsal funiculus of the rat spinal cord wereexamined 1 and 7 days after unilateral transection(rhizotomy) of the L4 and L5 dorsal roots, using lightand electron microscopy as well as X-ray microanaly-sis. Changes were observed only in the dorsal funicu-lus on the side of injury and included disintegration ofthe axonal cytoskeleton, enlargement of axonal mito-chondria, and widening of the myelin lamellae of theinjured axons. X-ray microanalysis demonstrated asignificant increase in intraaxonal sodium at 1 dayafter injury. This increase was abolished at 7 days, butat this stage there was a significant lowering of potas-sium in axons and myelin sheaths and of phosphorusin myelin as well as a marked increase in calcium inthe axoplasm of the degenerating axons. The nonneu-ronal cell compartment, largely composed of astro-cytes, showed elevated sodium, chlorine, and calciumand lowered potassium levels. The changes in chemi-cal composition paralleled an increase in immunore-activity for the calcium-binding Mts1 (S100A4) pro-tein, which is exclusively expressed by white matterastrocytes. The influx of calcium is likely to play acrucial role in the loss of axonal integrity after rhizo-tomy, while the alterations in potassium, and perhapsalso phosphorus, may contribute to activation of thenonneuronal cells, including the up-regulation of Mts1expression in astrocytes. © 2000 Academic Press

Key Words: nerve injury; spinal cord; morphology;ionic changes; X-ray microanalysis; Mts1; astrocytes;neuroplasticity.

INTRODUCTION

Injury to nerve fibers (axotomy) results in a processtermed Wallerian degeneration. The affected axon dis-integrates and, secondary to this, the surrounding my-elin sheath collapses and fragments. This process isaccompanied by prominent changes in the nonneuro-nal environment. In the central nervous system (CNS)

2360014-4886/00 $35.00Copyright © 2000 by Academic PressAll rights of reproduction in any form reserved.

astrocytes proliferate and hypertrophy, and microgliaproliferate and gradually transform into phagocytes(for references see 2).

Since neurons are highly excitable cells, proper func-tioning of the CNS requires an exceptionally preciseregulation of the ionic homeostasis in both intracellu-lar and extracellular compartments. In the extracellu-lar milieu this process is primarily associated withastrocytes, which form a functional “epithelium” withwhich the neurons continuously interact. Changes inthe ionic balance caused by axonal injury can thereforebe expected to markedly influence astrocytic functions.One of the ions which is likely to be the target ofastrocytic regulation is calcium (23). Astrocytesthroughout the CNS normally express high levels ofthe calcium-binding protein S100B, which has beenimplicated in several CNS pathologies, including, e.g.,Alzheimer’s disease (10). A recent study has demon-strated that astrocytes residing in the white matterexpress another calcium-binding protein belonging tothe S100 family as well, the metastasis-associatedMts1 protein (also termed S100A4) (11). Furthermore,the expression of this protein is markedly up-regulatedin spinal cord white matter astrocytes after injury toperipheral sensory nerve axons or to dorsal root axons.

Studies on peripheral nerve models have shown thattransection or severe compression (crush) of axons re-sults in rapid changes in the ionic composition of theaxoplasm and the surrounding myelin sheath distal tothe lesion. Ionic redistribution at the injury site willcause the resting membrane potential to fall, which islikely to spread rapidly in a somatofugal direction.Attempts by the axon to restore the normal restingmembrane potential in a situation where continuitywith the cell body is interrupted may cause severeenergy requirements, gradually leading to energy fail-ure and further deterioration of membrane function-ing, including the ability to counteract water influx.These changes will presumably trigger further down-ward directed changes within the axoplasm, eventually

237MORPHOLOGY AND CHEMICAL COMPOSITION OF DORSAL FUNICULUS

leading to a situation from which the injured axon andits myelin sheath cannot recover. In addition, alter-ations in, e.g., extracellular potassium and calciummay have a pronounced influence on the functionalstate of astrocytes and microglia in the vicinity of thedegenerating nerve fibers. However, there is no infor-mation on the magnitude or cellular or extracellularlocalization of ionic changes accompanying Walleriandegeneration in the CNS.

Transection of dorsal roots results in degeneration ofmyelinated axons ascending in the ipsilateral dorsalfuniculus. The first changes in the dorsal funiculusafter dorsal root injury have been reported to occuralready about 30 h after injury in 10-week-old rats (8),and 1 week after injury disintegration of axons and thenonneuronal responses are well under way. The mor-phological features of these processes have been exten-sively studied, but the molecular mechanisms under-lying the disruption of the injured nerve fibers and theactivation of nonneuronal cells are poorly known. Wehave applied X-ray microanalysis, which allows selec-tive determination of the chemical (elemental) compo-sition of cells and organelles, to identify changes inionic levels of degenerating axons and myelin sheathsto the dorsal root injury model, and combined thisanalysis with morphological and immunohistochemicalobservations, paying particular attention to changes inthe distribution of astrocytes expressing the calcium-binding Mts1 (S100A4) protein.

MATERIAL AND METHODS

Material

Twelve 8- to 10-week-old, female, Sprague–Dawleyrats (160–180 g body wt) were used for the study. Allexperiments were approved by the regional animalresearch ethics committee and carried out in accor-dance with the guidelines of the Society for Neuro-science. Prior to surgery and perfusion, animals wereanesthetized with chloral hydrate (35 mg/kg body wtintraperitoneally).

Control Material

Analyses obtained from the nonaffected, contralat-eral side served as internal control.

Surgery and Tissue Preparation

Nine animals were subjected to unilateral dorsalroot transection (rhizotomy). A midline incision wasmade and lumbar spine muscles removed from the L4and L5 vertebrae. The left lumbar dorsal roots L4 andL5 were exposed via a partial laminectomy andtransected close to the corresponding dorsal root gan-glia. After postoperative survival times of 1 and 7 daysthe animals were reanesthetized, and the region of

spinal cord segments L4 and L5 was rapidly exposedand processed for conventional and analytical electronmicroscopy (see below). These preparations were donewith the utmost care to avoid changes caused by han-dling tissue specimens.

Preparation for Electron Microscopy andMicroanalysis

Each spinal cord specimen was divided into twoparts and processed separately for conventional andanalytical electron microscopy. For microanalyticaland immunohistochemical purposes one part of thespinal cord was immediately cryofixed in propanecooled by liquid nitrogen (LN2). Serial 4- to 10-mm-thick cryosections were cut on a cryostat operating at230°C. Thinner sections aimed for X-ray microanalysiswere placed on specially designed specimen holders forX-ray microanalysis and freeze-dried. A thin carbonlayer was evaporated onto the samples prior to X-raymicroanalysis as previously described (28). Thickersections were used for histochemical and immunohis-tochemical staining. The second part of the spinal cordwas fixed in 2.5% glutaraldehyde in cacodylate buffer,pH 7.2, postfixed in 1% osmium tetroxide in the samebuffer, and embedded in Agar 100 epoxy resin (AgarScientific, Stansted, UK) for conventional electron mi-croscopy. Control material was fixed in the same wayas material from injured animals.

Immunohistochemistry

Cryostat sections parallel to those used for micro-analysis were air-dried and rinsed in phosphate-buff-ered saline (5–10 min) prior to incubation in bovineserum albumin and 0.3% Triton X-100 (Sigma, St.Louis, MO) for 1 h at room temperature. Sections wereincubated overnight at 4°C in antibodies to Mts1 (rab-bit polyclonal, 1:1000, a gift from Dr. E. Lukanidin,Copenhagen, Denmark). The immune complex was vi-sualized with fluorescein-conjugated sheep anti-rabbitIgG (Jackson, West Grove, PA; 1:40). Sections wereviewed and photographed in a Nikon Eclipse fluores-cence microscope. These microphotographs served asguides for localizing astrocytes in the degeneratingdorsal funiculus while searching for astrocytes duringthe microanalytical procedure.

X-Ray Microanalysis and Electron Microscopy

The specimens were viewed and analyzed in thescanning-transmission (STEM) mode of the electronmicroscope (Hitachi H7100, Tokyo, Japan) at an accel-erating voltage of 100 kV. X-ray microanalysis wasperformed with an Oxford Instruments ISIS energy-dispersive spectrometer system (Oxford, UK). The res-olution of analysis (sampling size) was approximately 2mm, unless otherwise indicated. The specimens were

ca

238 WROBLEWSKI, ROOMANS, AND KOZLOVA

examined in the scanning-transmission mode. The to-tal counting time (lifetime) was 50 s. Absolute concen-trations were calculated using standards and ex-pressed in mmol/kg dry wt (19, 25, 28).

Statistical Analysis

Student’s t test was used to evaluate statistical sig-nificance in elemental content in the injured group andthe controls. P values less than or equal to 0.05 wereconsidered statistically significant.

RESULTS

Light Microscopy

Light microscopy was performed on 4- to 6-mm-thickryosections adjacent to the ones used for X-ray micro-nalysis as well as on 1-mm-thick sections from an

FIG. 1. Low-power scanning-transmission electron image of a 4specimen was taken 7 days after injury. It is possible to identify porthe degenerated area on the left side of the midline which runs alonopposite side. (Inset) An adjacent section showing the distribution ofthe degenerated dorsal funiculus. The corresponding area was used

adjacent part of the spinal cord, which was aldehyde-and osmium-fixed and plastic-embedded for subse-quent morphological examination in the electron mi-croscope. The zone of degeneration in the left dorsalfuniculus was readily distinguished due to the pres-ence of myelinated axons with abnormal appearance(Fig. 1). In cryosections this zone was also character-ized by expanded spaces between the axons, probablydue to extracellular ice crystal formation.

Electron Microscopy

One-micrometer-thick plastic sections viewed in thelight microscope were used as guides for the identifi-cation of areas of interest in the spinal cord and wereused both in conjunction with the examination of plas-tic-embedded ultrathin (60-nm) sections and for theorientation and microanalysis of the 4- to 6-mm-thickcryosections.

cryosection of spinal cord on a pure carbon specimen holder. Thens of the dorsal funiculus (DF) and the dorsal horn (DH) as well ashe rupture in the section and the area of normal appearance on thes1 protein (visualized with fluorescein). Mts1 was mostly present inmicroanalytical purposes. Bar, 100 mm.

-mmtiog tMtfor

toosm


Morphological Observations on CryosectionedMaterial Observed in the STEM Modeof Microscopy and in Adjacent SectionsObserved in the Light Microscope

The morphology in 4- to 6-mm cryosections as ob-served in the STEM mode of microscopy was adequateto visualize white and gray matter (Figs. 1 and 2) andalso to define degenerating versus nondegeneratingareas within the dorsal funiculus. In most cases themidline could be identified, which provided additionalorientation within the section. An accurate localizationof the white matter with degenerating versus normalaxons was especially valuable in specimens taken 1day after injury when unambiguous signs of morpho-logical changes were absent. Myelinated axons wereeasily identified and their myelin sheath and axoplasmwere readily distinguished in the degenerating as wellas unaffected white matter.

Seven days after injury a variety of morphologicalchanges were observed in the degenerating zone. My-elinated nerve fibers frequently appeared obliquely orpartly longitudinally cut while fibers in the unaffectedareas were almost invariably cut transversely and dis-played a more regular, round appearance. Myelinsheaths in the zone of degeneration were often irregu-lar and the myelin lamellae were more loosely wrappedaround the axon. The axoplasm of degenerating fiberswas often divided into several strands within the sur-rounding myelin and contained dense round particles,presumably mitochondria or secondary lysosomes andbundles of cytoskeletal components.

FIG. 2. (a) Low-power scanning-transmission electron image of ahe L4 and L5 dorsal roots. The section is placed on a plastic film (bserved only in the dorsal funiculus on the side of injury on the leftbserved on the opposite side (AN). The amorphous area on the lowpecimens. Bar, 50 mm. (b) High-power scanning-transmission electryelin (my) surrounds the axoplasm (axo). Several mitochondria wi

Morphological Observations on ConventionallyPrepared Material for TransmissionElectron Microscopy

Degenerative changes were easily identified 7 daysafter injury while no morphological changes were ob-served in specimens taken 1 day after injury (Fig. 3).Myelin as well as axoplasm was affected. Myelinchanges included an irregular outline and an increaseddistance between single lamellae in the myelin sheath.Axons in the degenerating zone displayed variablechanges. Many axons showed loss of neurofilamentsand microtubuli. In more advanced cases, the axo-plasm displayed increased electron density, and theaxon appeared shrunken. Mitochondria were fre-quently enlarged and sometimes contained dense gran-ules. Vesicular profiles and unidentified “debris” werepresent in the axoplasm of some axons. Glial cellperikarya were occasionally observed among the de-generating nerve fibers. Numerous astrocytic pro-cesses containing bundles of typical astrocytic interme-diate filaments intermingled with the degeneratingnerve fibers.

Immunocytochemistry

Using antibodies against the calcium-binding pro-tein Mts1 on cryosections, weak positive reaction prod-uct was found throughout the white matter of the spi-nal cord. The degenerating zone of the dorsal funiculusshowed a moderate increase at 1 day after rhizotomyand a marked up-regulation 7 days after injury, when

mm cryosection of spinal cord, 7 days after unilateral transection ofn a pure carbon specimen holder (SH). Morphological changes aree (AD) of the midline (- - -), while no changes in axon structure are

eft side of the picture represents the embedding medium for frozenimage of a single axon from the injured side of white matter. Densedense granules (dg) are present within the axoplasm. Bar, 0.5 mm.

4-*) osid

er lonth


large numbers of heavily labeled cell profiles werefound (Fig. 1, inset). In a previous study Mts1 immu-noreactivity was shown to be present exclusively inwhite matter astrocytes and their processes (cf. 11).

X-Ray Microanalysis

One day after injury. Since axons and myelinsheaths had a normal morphology 1 day after injury,when viewed in the scanning-transmission electron mi-croscope, the localization of the proper zones in theright and left dorsal funiculi was based on matchingthe correct topographic area in the light microscopicsections to the corresponding area in the cryosectionsfor microanalysis. Astrocytic cell bodies and processeswere measured if they could be identified on adjacentsections stained with antibodies to Mts1. The mostprominent finding was an increase in sodium in axonsand myelin to almost twofold that of control areas. Inaddition, there was a trend toward lower potassiumand higher calcium and chlorine levels in axons andmyelin, but these changes were not statistically signif-icant (data not shown).

FIG. 3. Transmission electron micrographs from the degeneratichanges are in the axoplasm in one of the axon (arrow). There are alrandom distributed changes among the axons. Some of the axonsneurofilament and microtubuli were absent. (b) Single astrocyte cell wAn enlarged mitochondrion (arrow) is present in the axoplasm of thefilaments. Bar, 1 mm.

Seven days after injury. The overall analysis of thenormal and degenerating dorsal funiculus covered ar-eas containing hundreds of cells (diameter of the ana-lyzed area 30–50 mm). The most striking finding was alow potassium level in the degenerating white matter.In addition, the calcium level was somewhat elevated.

The most striking changes at the subcellular levelwere a lowered potassium and an increased calciumcontent in both axoplasm and myelin in the degener-ating white matter (Figs. 4a–4d, 5, and 6). The highestlevels of calcium were found in the dense structurespresent in the axoplasm and which were presumablymitochondria. These structures also showed elevatedphosphorus levels compared to the rest of the axo-plasm. In the degenerating white matter potassiumwas fourfold lower and calcium approximately fourfoldhigher. Phosphorus was significantly lower in degen-erating myelin in the white matter. The sulfur contentin the axoplasm was stable at approximately 140mmol/kg dry wt, but a significant loss of sulfur wasnoted in the myelin. Reactive astrocytes, identified bytheir immunoreactivity for Mts1 in adjacent sections,showed relatively high sodium values, but values for

dorsal funiculus 7 days after dorsal rhizotomy. (a) Most prominentchanges of the shape of the entire myelin sheath. Axoplasm showedd both microtubuli and neurofilaments intact but in many axonsnuclei and prominent cytoplasmic filaments lay between the axons.

n. (c) Three astrocyte cell extensions (A) with prominent cytoplasmic

ngsohaith

axo


chloride and calcium were within the normal range(Fig. 7).

DISCUSSION

Our findings show that the early phase of Walleriandegeneration in the dorsal funiculus of the spinal cordwhite matter is accompanied by marked alterations inseveral elements which play a crucial role for normalcellular function. These alterations appear to occuralready at 1 day after dorsal root injury and becamevery distinct at 7 days postoperatively. At this stage,the degenerating areas within the white mattershowed clear-cut morphological alterations as well. Im-portantly, the lesions inducing these responses weremade several centimeters from the spinal cord seg-

FIG. 4. Examples of different elemental spectra in white matter:after 7 days degeneration; (d) mitochondrial dense inclusion aftersymbols—sodium (Na), phosphorus (P), sulfur (S), chlorine (Cl), pot

ments analyzed and without interfering with spinalcord blood supply, thereby excluding, e.g., ionicchanges due to ischemia, which in itself can induceaxon degeneration (24).

Primary microanalytical results were obtained fromlarge areas (diameter 30–50 mm) containing hundredsof cells in the normal and degenerating white matter.These analyses included information on both extracel-lular and intracellular contents. All cell types repre-sented in the white matter, including blood capillaries,astrocytes, and axons, were simultaneously analyzed.This type of overall analysis might be of interest asadditional recourse for proper area identification be-sides the optical recognition prior to analysis on thecellular and subcellular levels. The analysis performedon the single-cell level gave more informative data,

normal axoplasm; (b) axoplasm after 7 days degeneration; (c) myelindays degeneration. The elements are marked with their chemicalium (K), and calcium (Ca).

(a)7

ass

a

7(cca


however, compared with “overall” analysis. Analyzingcomponents of degenerating and unaffected white mat-ter within the same section is an advantage since allanalytical parameters are the same. However, this wasnot feasible with astrocytes, because it was difficult toidentify these cells in control areas in cryosections. Inthe degenerating area, reactive astrocytes were identi-fied by the use of parallel cryosections immunostainedwith antibodies to Mts1, a marker exclusively ex-pressed by white matter astrocytes and strongly up-regulated after dorsal root injury (11).

Previous data on subcellular and molecular changesduring Wallerian degeneration are almost exclusivelyderived from studies on transected or crushed periph-eral nerve. Schlaepfer and Micko (22) found that the

FIG. 5. Elemental content (mean and SD) in the axoplasm (axo)days after axotomy. Concentrations of sodium (Na), magnesium

Mg), phosphorus (P), sulfur (S), chlorine (Cl), potassium (K), andalcium (Ca) in the control area (H) and the degenerated (D) area areompared. The significant differences (P , 0.05) are marked with ansterisk.

microtubule network in the axoplasm is the first com-ponent to degenerate, followed by neurofilaments. Fi-nally, myelin sheaths collapse and undergo fragmen-tation. A similar morphological process appears to oc-cur during Wallerian degeneration in the CNS,although with an attenuated time course (8). Our ul-trastructural findings on immersion-fixed, plastic-em-bedded sections are in good agreement with previousdata. In addition to cytoskeletal changes in the axons,and a loosening of the surrounding compact myelin,intraaxonal mitochondria were commonly enlarged

FIG. 6. Elemental content (mean and SD) in the myelin (my), 7days after axotomy. Concentrations of sodium (Na), magnesium(Mg), phosphorus (P), sulfur (S), chlorine (Cl), potassium (K), andcalcium (Ca) in the control area (H) and the degenerated (D) area arecompared. The significant differences (P , 0.05) are marked with ansterisk.

scBflanptcd

ifnipdnpwulel


and contained dense bodies. These mitochondrial alter-ations may reflect attempts to meet increasing energydemands associated with the difficulties in maintain-ing normal membrane function in the injured axons.

Electrophoretic analysis of nerve and myelin pro-teins in the degenerating peripheral nerve indicatedthat proteins forming neurofilaments and microtubuliwere broken down and lost in exchange for compoundswith a much lower molecular weight (22). Subse-quently, there was a selective loss of different myelinproteins starting on day 4 after nerve transection. Inour electron microscopic examination of the degenerat-ing white matter we observed numerous axons contain-ing various forms of debris, presumably largely origi-nating from disrupted structures and high molecularweight components of the axonal cytoskeleton. Thedecrease in phosphorus and sulfur in the myelin, asfound in the present study, fits in well with the notionof myelin disintegration.

The elemental changes in axons and myelin sheathsparalleled each other. One day after injury the onlysignificant change was an increase in sodium levels(data not shown). Similar findings were made in de-generating peripheral nerve during the first 2 daysafter injury (14). Axotomy may induce a massive injury

FIG. 7. Elemental content (mean and SD) in the astrocytes, 7days after axotomy. Concentrations of sodium (Na), magnesium(Mg), phosphorus (P), sulfur (S), chlorine (Cl), potassium (K), andcalcium (Ca) in the degenerating dorsal funiculus are given.

potential based on influx of sodium at the injury siteand propagating along the nerve fibers (at the sites ofthe nodes of Ranvier) into the dorsal funiculus. Thehigh sodium levels may be remains from this process.Alternatively, sodium ions could diffuse in and potas-sium ions could diffuse out at the site of damage.

At 1 week after dorsal rhizotomy several elementalcomponents were affected. We found markedly ele-vated calcium in degenerating nerve fibers in the dor-sal funiculus, indicating that the normal regulation ofcalcium influx into the axoplasm and myelin sheathwas disturbed. High calcium and phosphorus levelswere detected in dense intraaxoplasmic bodies, whichin the morphological analysis were found to representmitochondria containing dense granules. Such gran-ules have been found under different pathological con-ditions (27), as well as in some normal tissues (26, 27).It is likely that the axolemma of injured axons and thegradually disintegrating myelin membranes becomeincreasingly permeable to extracellular calcium ions.One way of counteracting an excessive accumulation ofcalcium in the cytosol is an increased uptake of calcium(and phosphorus) in mitochondria.

Our findings of an increase in calcium levels in in-jured axons are in line with previous reports implicat-ing calcium as a prime factor in Wallerian degenera-tion of peripheral nerve. In a series of in vitro experi-ments, elevated levels of calcium were found toactivate calcium-activated proteases (calpains) andsubsequent breakdown of neurofilament proteins,thereby causing a collapse of the “backbone” of theaxonal cytoskeleton (20, 21). These findings were con-firmed and extended by George et al. (7) who demon-trated that axon degeneration in vitro is mediated byalpains that cleave major axonal structural proteins.ased on the use of various agents which block calciumux across membranes, these authors suggested thatfter axotomy extracellular calcium might enter theerve cell via a specific ion transport mechanism. Arevious microanalytical study of Wallerian degenera-ion in peripheral nerve has also shown an increasedalcium concentration during the first 2 postinjuryays (14).Martinez and Ribeiro (15), using calcium histochem-

stry and microanalysis on injured peripheral nerve,ound the same distribution of calcium in controlerves and in normally appearing axons within the

njured nerve 30 h after axotomy. Injured axons withartially disrupted cytoskeleton had a greater inci-ence of calcium precipitates, in particular close toodes of Ranvier, as well as in mitochondria and endo-lasmic vesicles. Degenerating axons in our materialere mainly cross-sectioned, and we were thereforenable to obtain sufficient amount of data on elemental

evels in the nodal area. Since this area is directlyxposed to the extracellular compartment it seemsikely, however, that the influx of ions into the axo-


plasm would be more rapid and extensive at this sitethan along the internodal regions.

Microanalytical data from the CNS are limited toexamination of chemical changes during the first dayfollowing spinal cord trauma (9). The impact of thistrauma on nerve fibers, blood vessels, and nonneuronalcells will be markedly different from the process ofWallerian degeneration, and available data are there-fore not directly comparable. Lightly traumatizednerve fibers showed essentially normal chemical com-position, but in more extensively traumatized nervefibers there was a loss of potassium and an increase insodium, chlorine, and calcium.

In addition to playing a crucial role in the degener-ative processes in the injured nerve fibers, calcium mayalso act as a mediator of astrocytic activation (6, 16,18). X-ray microanalysis of astrocytes was somewhathampered by the difficulty in identifying astrocytes innondegenerating areas in the cryosections. To judgefrom the absolute values, there was no evidence ofcalcium accumulation in astrocytes. However, signifi-cant calcium influx may have been rapidly “buffered”in the astrocytic syncytium or occurred between thetime points examined (1 and 7 days). Astrocytic prolif-eration occurs mainly between 2 and 4 days after rhi-zotomy (13), and up-regulation of markers for astro-cytic activation is prominent between 2 and 7 days (12).This up-regulation includes Mts1 (S100A4) protein(11), which is a calcium-binding protein implicated intumorigenesis, cell motility, and neurite outgrowth.Although the role of this protein with regard to nervedamage is not clear, it can be speculated that its cal-cium-buffering properties contribute to a lowering ofthe calcium concentrations in the fluid surrounding theaxon in contact with the astrocyte. This would assistthe damaged axon (which has little calcium-bufferingcapacity) in reducing its intracellular calcium concen-tration, and this could contribute to limiting or slowingdown the process of nerve degeneration.

Parallel to the elevated levels of calcium within ax-ons and myelin sheaths, there was a marked loss ofpotassium from these compartments at 7 days afterrhizotomy. This is in line with previous findings ondegenerating peripheral nerve (14). Since there wasno significant increase in sodium levels the Na/K-ATPases are presumably intact. Moreover, this disso-ciation between relatively unaffected sodium levelsand a marked loss of potassium suggests that changesin axonal membrane functions are to some extent se-lective. The loss of potassium may be the result ofimpaired function of other membrane proteins than theNa/K-ATPase. These changes may cause voltage-gatedor potassium leak channels to remain in an open state,resulting in further potassium efflux. Axon injury maycause disintegration of potassium-binding proteins,thereby impairing the maintenance of the normallyhigh intracellular potassium levels. Being the most

common cation in the cell, potassium has a unique rolein maintaining the resting membrane potential. Lossof potassium might also be connected with loss of neg-atively charged groups such as phosphate and/or car-boxyl groups (5). At the same time, low phosphorus(phosphate) levels in degenerating axoplasm mightpartly reduce the difference in positive and negativeions and macromolecules in the injured cell.

One of the major functions of astrocytes is to regulatepotassium homeostasis (17). Given the significant lossof potassium in degenerating axons and myelinsheaths, a build-up of extracellular potassium levelswould be expected. Since X-ray microanalysis revealedno changes in potassium levels in reactive astrocytes,the excess extracellular potassium is most probablyrapidly taken up and equilibrated via the network cre-ated via the gap junctions of the astrocytic syncytium.At the same time, a transient increase in extracellularpotassium levels may serve as an efficient stimulus foractivating astrocytes (see, e.g., 1), similar to the rolethat sodium may have in tumor cells (3, 4).

CONCLUSIONS

The present study has identified major changes insodium, potassium, phosphorus, and calcium in my-elinated nerve fibers undergoing Wallerian degenera-tion in the spinal cord. These findings are largely con-sistent with previous data from degenerating periph-eral nerve. Lowering calcium levels in the extracellularcompartment or inhibiting calcium influx into dam-aged peripheral nerve fibers has been shown to mark-edly attenuate disintegration of the injured axons. Theregulation of extra- and intracellular calcium levels,e.g., via calcium-binding proteins such as Mts1, maytherefore play a crucial role in CNS repair processes.

ACKNOWLEDGMENTS

The technical assistance of Mr. Anders Ahlander and Ms. Ing-Marie Olsson is gratefully acknowledged. This study was supportedby the Swedish Medical Research Council, Project 5420.

REFERENCES

1. Abiru, Y., R. Katoh-Semba, C, Nishio, and H. Hatanaka. 1998.High potassium enhances secretion of neurotrophic factors fromcultured astrocytes. Brain Res. 809: 115–126.

2. Aldskogius, H., and E. N. Kozlova. 1998. Central neuron–glialand glial–glial interactions following axon injury. Prog. Neuro-biol. 55: 1–26.

3. Cameron, I. L., and N. K. R. Smith. 1989. Applications ofelectron probe X-ray microanalysis to the study of ionic regula-tion of growth of normal and cancer cells. In Microprobe Anal-ysis in Medicine (P. Ingram, J. D. Shelburne, and V. L. Roggli,Eds.), pp. 291–301. Hemisphere, Washington, DC/New York.

4. Cameron, I. L., N. K. R. Smith, T. B. Pool, and R. L. Sparks.1980. Intracellular concentration of sodium and other elements

1

1

1

1

1

1

1

1

1

1

2

2


as related to mitogenesis and oncogenesis in vivo. Cancer Res.40: 1493–1500.

5. Cameron, I. L., W. E. Hardman, K. E. Hunter, C. Haskin,N. K. R. Smith, and G. D. Fullerton. 1990. Evidence that amajor portion of cellular potassium is “bound.” Scanning Mi-crosc. 1: 165–178.

6. Du, S., A. Rubin, S. Klepper, S. Barrett, Y. C. Kim, H. W. Rhim,E. B. Lee, C. W. Park, G. J. Markelonis, and T. H. Oh. 1999.Calcium influx and activation of calpain I mediate acute reac-tive gliosis in injured spinal cord. Exp. Neurol. 157: 96–105.

7. George, E. B., J. D. Glass, and J. W. Griffin. 1995. Axotomyinduced axonal degeneration is mediated by calcium influxthrough ion-specific channels. J. Neurosci. 15: 6445–6452.

8. George, E. B., and J. W. Griffin. 1994. The proximo-distalspread of axonal degeneration in the dorsal columns of the rat.J. Neurocytol. 23: 657–667.

9. Greene, W. B., and L. G. Walsh. 1994. Cryo-jet preservation ofcalcium in the rat spinal cord. Scanning Microsc. 3: 587–600.

0. Griffin, W. S., J. G. Sheng, J. E. McKenzie, M. C. Royston, S. M.Gentleman, R. A. Brumback, L. C. Cork, M. R. Del Bigio, G. W.Roberts, and R. E. Mrak. 1998. Life-long overexpression ofS100beta in Down’s syndrome: Implications for Alzheimerpathogenesis. Neurobiol. Aging 19: 401–405.

1. Kozlova, E. N., and E. Lukanidin. 1999. Metastasis-associatedMts1 (S100A4) protein is selectively expressed by white matterastrocytes and is up-regulated after peripheral nerve or dorsalroot injury. Glia 27: 249–258.

2. Liu, L., J. K. E. Persson, M. Svensson, and H. Aldskogius. 1998.Glial cell responses, complement and clusterin in the centralnervous system following dorsal root transection. Glia 23: 221–238.

3. Liu, L., M. Rudin, and E. N. Kozlova. 1999. Glial cell prolifer-ation and dorsal root transitional zone modification after dorsalrhizotomy or sciatic nerve transection in the adult rat. Exp.Brain Res., in press.

4. LoPachin, R. M., C. M. Castiglia, and A. J. Saubermann. 1990.Effects of axotomy on distribution and concentration of ele-ments in rat sciatic nerve. J. Neurochem. 54: 320–332.

5. Martinez, A. M., and L. C. Ribeiro. 1998. Ultrastructural local-ization of calcium in peripheral nerve fibres undergoing Walle-rian degeneration: An oxalate-pyroantimonate and X-ray mi-croanalysis study. J. Submicrosc. Cytol. Pathol. 30: 451–458.

6. Nedergaard, M. 1994. Direct signaling from astrocyte to neu-rons in cultures of mammalian brain cells. Science 263: 1768–1771.

7. Newman, E. A. 1995. Glial cell regulation of extracellular po-tassium. In Neuroglia (H. Kettenmann and B. S. Ransom,Eds.), pp. 717–731. Oxford Univ. Press, New York.

8. Parpura, V., F. Basarsky, K. Liu, S. Jeftinija, P. G. Jeftinija,and P. G. Haydon. 1994. Glutamate-mediated astrocyte–neuronsignaling. Nature 369: 744–747.

9. Roomans, G. M. 1988. Quantitative X-ray microanalysis of bi-ological specimens. J. Electron Microsc. Tech. 9: 19–44

20. Schlaepfer, W. W. 1971. Experimental alterations of neurofila-ments and neurotubules by calcium and other ions. Exp. CellRes. 67: 73–80.

21. Schlaepfer, W. W., and M. B. Hasler. 1979. Characterization ofthe calcium-induced disruption of neurofilamentes in rat pe-ripheral nerves. Brain Res. 168: 299–309.

22. Schlaepfer, W. W., and S. Micko. 1978. Chemical and structuralchanges of neurofilaments in transected rat sciatic nerve.J. Cell Biol. 78: 369–378.

23. Verkhratsky, A., and H. Kettenmann. 1996. Calcium signalingin glial cells. Trends Neurosci. 19: 346–352.

24. Waxman, S., B. R. Ransom, and P. K. Stys. 1991. Non-synapticmechanisms of Ca11 mediated injury in CNS white matter.Trends Neurosci. 14: 461–468.

25. Wroblewski, J., R. M. Muller, R. Wroblewski, and G. M. Room-ans. 1983. Quantitative X-ray microanalysis of semi-thick cryo-sections. Histochemistry 77, 447–463.

6. Wroblewski J., R. Wroblewski, C. Mory, and C. Colliex. 1991.Elemental analysis and fine structure of mitochondrial gran-ules in growth plate chondrocytes studied by electron energyloss spectrometry and energy dispersive X-ray microanalysis.Scanning Microsc. 5: 885–894.

27. Wroblewski, R., M. Anniko, and L. Edstrom. 1981. A new typeof striated muscle in mammalian body: Morphological, histo-chemical and X-ray microanalytical observations of stapediusmuscle in guinea pig. J. Ultrastruct. Res. 76: 46–56.

8. Wroblewski, R., G. M. Roomans, E. Jansson, and L. Edstrom.1978. Electron probe X-ray microanalysis of human musclebiopsies. Histochemistry 55: 281–292.

Effects of Dorsal Root Transection on Morphology and Chemical Composition of Degenerating Nerve...

Documents

Transcript of Effects of Dorsal Root Transection on Morphology and Chemical Composition of Degenerating Nerve...