Brain, Vol. 123, No. 3, 519-531,
March 2000
© 2000 Oxford University Press
Invited review |
Axonal loss results in spinal cord atrophy, electrophysiological abnormalities and neurological deficits following demyelination in a chronic inflammatory model of multiple sclerosis
1 Molecular Neuroscience Program and Departments of 2 Neurology and 3 Immunology, Mayo Clinic and Foundation, Rochester, Minnesota, USA
Correspondence to:
Moses Rodriguez, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA E-mail: rodriguez.moses{at}mayo.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMethodResultsDiscussionReferences |
|---|
Recent pathological studies have re-emphasized that axonal injury is present in patients with multiple sclerosis, the most common demyelinating disease of the CNS in humans. However, the temporal profile of demyelination and axonal loss in multiple sclerosis patients and their independent contributions to clinical and electrophysiological abnormalities are not completely understood. In this study, we used the Theiler's murine encephalomyelitis virus model of progressive CNS inflammatory demyelination to demonstrate that demyelination in the spinal cord is followed by a loss of medium to large myelinated fibres. By measuring spinal cord areas, motor-evoked potentials, and motor coordination and balance, we determined that axonal loss following demyelination was associated with electrophysiological abnormalities and correlated strongly with reduced motor coordination and spinal cord atrophy. These findings demonstrate that axonal loss can follow primary, immune-mediated demyelination in the CNS and that the severity of axonal loss correlates almost perfectly with the degree of spinal cord atrophy and neurological deficits.
rotarod; Theiler's virus; conduction; inflammation; neuropathology
ANOVA = analysis of variance; d.p.i. = days post-infection; EDSS = expanded disability status score; MEP = motor-evoked potential; MRS = magnetic resonance spectroscopy; NAA = N-acetylaspartate; TMEV = Theiler's murine encephalomyelitis virus
|
Introduction |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Multiple sclerosis is the most common demyelinating disease of the CNS in humans and is characterized by neurological deficits, including impaired vision, sensory deficits, motor weakness, intention tremor and ataxia. It is widely accepted that the most common pathological abnormality, inflammatory demyelination in the CNS, contributes to the functional impairment in patients (Adams et al., 1997
)
Recently, detailed pathological studies have thoroughly characterized the presence of axonal damage (Ferguson et al., 1997; Trapp et al., 1998) mentioned in Charcot's early description of multiple sclerosis (Charcot, 1868). These studies are supported by an observed decrease in N-acetyl groups [primarily N-acetylaspartate (NAA)] by magnetic resonance spectroscopy (MRS) in the brains of multiple sclerosis patients (Davie et al., 1995, 1997; De Stefano et al., 1997, 1998; Narayanan et al., 1997; Fu et al., 1998). As NAA is found almost exclusively in neurons (Urenjak et al., 1993), it is assumed that reductions in NAA represent axonal loss. Several investigators have also described a correlation between NAA reductions and worsening of the expanded disability status scores (EDSS) (Kurtzke, 1983) in patients (Davie et al., 1997; De Stefano et al., 1997, 1998; Fu et al., 1998).
The recent emphasis on axonal damage in multiple sclerosis has led to a more detailed evaluation of the underlying parameters that contribute to neurological deficits. It has long been noted that damage in the spinal cord is important in the development of neurological dysfunction of multiple sclerosis patients (Oppenheimer, 1978). Kidd and colleagues described spinal cord involvement in 74% of patients (Kidd et al., 1993). However, MRI has also demonstrated that spinal cord atrophy is a prominent feature of multiple sclerosis and correlates well with disability (Kidd et al., 1993, 1996; Filippi et al., 1996; Losseff et al., 1996; Lycklama a Nijeholt et al., 1998; Stevenson et al., 1998). It is assumed that this spinal cord atrophy represents a marker of axonal loss.
The aforementioned studies demonstrate that both demyelination and axonal loss may contribute independently to disability in multiple sclerosis patients. However, the temporal profile of these events and their distinct contributions to electrophysiological abnormalities and functional deficits are not completely understood. We have therefore used a viral model of progressive inflammatory demyelination to determine how these two factors are related to both neurological and electrophysiological abnormalities.
Intracerebral injection of Theiler's murine encephalomyelitis virus (TMEV) into susceptible strains of mice results in pathological abnormalities and neurological deficits similar to those seen in multiple sclerosis (Lehrich et al., 1976; Dal Canto and Lipton, 1977, 1979; Lipton and Dal Canto, 1976a; Rodriguez et al., 1987). This biphasic disease is characterized by early infection of the grey matter, which is cleared within 21 days, followed by viral persistence in the spinal cord white matter (Lipton, 1975). Virus antigen has not been detected in neurons during the chronic stage of disease (following 21 days) and should not contribute directly to progressive neuronal disruption (Dal Canto and Lipton, 1982; Rodriguez et al., 1983); however, limited neuronal disruption may occur during the acute phase of the disease. White matter demyelination in the spinal cord is immune-mediated (Lipton and Dal Canto, 1976b), which may be relevant to the study of multiple sclerosis.
In the present study, we assessed frequency distributions of axons in normally myelinated areas at various stages of disease in the spinal cord. We assessed axonal loss from the normal appearing white matter to avoid confounding variables such as oedema and inflammation that complicate analysis of demyelinated lesions. Previous multiple sclerosis studies have also focused on analysing the normal appearing white matter (Davie et al., 1995, 1997; De Stefano et al., 1997, 1998; Narayanan et al., 1997; Fu et al., 1998). We hypothesized that axonal loss would be apparent in the normally myelinated regions and that the severity of axonal loss would predict the degree of spinal cord atrophy, electrophysiological abnormalities and neurological deficits (all parameters assessed in multiple sclerosis patients). We also determined the extent and severity of spinal cord demyelination (a measure of lesion load) to determine how demyelination contributed to the disease process.
|
Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Mice
The following mouse strains were purchased from The Jackson Laboratories (Bar Harbor, Me., USA): C57BL/10 (n = 8) (prototypic resistant strain—H-2b) and SJL/J (n = 75) (prototypic susceptible strain—H-2s). Mice were anaesthetized and intracerebrally injected at 8 weeks of age with 2 x 106 p.f.u. (plaque-forming units) of the Daniel's strain of TMEV in a 10 µl volume. Age-matched mice were intracerebrally sham-infected with 10 µl of PBS (phosphate-buffered saline) to serve as controls for the functional assays. [A previous report describing methodology used to quantify spinal cord demyelination, remyelination, atrophy and axonal loss in detail used seven infected and seven sham-infected mice from this study (McGavern et al., 1999a
).] Care and handling of mice conformed to the guidelines of both the National Institutes of Health and Mayo Clinic Animal Care and Use Committee.
Tissue processing
All analyses (which included calculation of spinal cord demyelination, spinal cord atrophy and axonal area distributions) were blinded and conducted on tissue processed in the following manner. Mice were anaesthetized with an intraperitoneal injection of 10 mg of pentobarbital and perfused via intracardiac puncture with Trump's fixative (phosphate-buffered 4% formaldehyde with 1% glutaraldehyde; pH 7.2). The spinal cord was then removed, sectioned coronally into 1 mm blocks, post-fixed with osmium tetroxide and embedded in Araldite (Polysciences, Warrington, Penn., USA). One micrometre thick cross-sections were cut from every third 1 mm block, resulting in the analysis of a minimum of 10 blocks. These 1 µm cross-sections were stained with 4% paraphenylenediamine to label myelin. For the serial section experiment described only in the motor function segment of the results, spinal cords were processed as described above; however, every other 1 mm block was embedded and cut to obtain a more complete representation of all cord regions. An average of 15 sections per mouse was analysed.
Analysis of spinal cord demyelination
To calculate the extent of spinal cord demyelination in susceptible strains of mice infected with TMEV, a Zeiss interactive digital analysis system (ZIDAS) and camera lucida attached to a Zeiss photomicroscope (Carl Zeiss Inc., Thornwood, NY, USA) were used as reported (McGavern et al., 1999a). Ten to twelve 1 µm thick transverse spinal cord sections per mouse were prepared as described above. The percentage of spinal cord demyelination per mouse was calculated by first determining the total white matter area for all spinal cord sections by manually tracing the regions. Next the area of spinal cord demyelination was determined by manually tracing each of the demyelinating lesions. Areas of demyelination were defined as containing naked axons, macrophage infiltration, myelin ovoids and degenerated axon profiles. The percentage demyelination per mouse was obtained by dividing the total area of demyelination by the total area of white matter sampled. These data were plotted for all susceptible mice and statistically compared using a one-way ANOVA (analysis of variance). Pairwise comparisons were performed using the Student–Newman–Keuls method (P < 0.05). Correlation coefficients between percentages of spinal cord demyelination and rotarod performances were calculated using the Pearson product moment correlation (P < 0.05).
Spinal cord atrophy
The percentages of spinal cord atrophy were determined using the tissue described above. An Olympus Provis AX70 microscope fitted with a SPOT colour digital camera and a x1.25 objective was used to digitize all spinal cord cross-sections cut for each mouse as described (McGavern et al., 1999a). A program written for the KS400 image analysis software (Kontron Elektronik GmbH, Munich, Germany) was then used to calculate the total cord area, grey matter area, posterior white matter area and the remaining anterior and lateral white matter area (which included the anterior, lateral and anterolateral columns) after manually outlining the regions. The cord sections were classified into one of three categories based on their location within the cord: C1–C7, C8–T11 and T12/13–L3. The data for all infected mice were then plotted as a percentage change from a group of uninfected SJL/J mice (n = 7) that showed no spinal cord abnormalities. Statistical comparisons between the original spinal cord area measurements were performed using a one-way ANOVA. Pairwise comparisons were done using the Student–Newman–Keuls method (P < 0.05). Correlation coefficients between C7 lateral/anterior column areas and rotarod performances were calculated using the Pearson product moment correlation (P < 0.05). (One mouse was excluded from the correlation because the C7 spinal cord section was not available.)
Myelinated axonal area distributions
Calculation of axonal area distributions was performed as reported (McGavern et al., 1999a) on the paraphenylenediamine-stained cross-sections described above. The sections used for the study were cut precisely at 1 µm and stained for exactly 20 min with the same batch of 4% paraphenylenediamine to ensure an identical intensity of myelin labelling. For each animal, an Olympus Provis AX70 microscope fitted with a SPOT colour digital camera and a x60 oil objective was used to digitize sample images from the smallest thoracic spinal cord section (T6) according to the sampling scheme shown in Fig. 3E. Images were captured from regions containing a relative absence of demyelination to ensure the quantitation of only myelinated fibres. Approximately 145 000 µm2 of white matter was sampled from the T6 section of each animal. Using a program written for the KS400 image analysis software, myelinated axonal area distributions were calculated following segmentation of the grey values (145–255) corresponding to the axoplasm from each image. The program automatically calculated the area of each axon in the field from the segmented binary image after regions corresponding to the vasculature, cell bodies, longitudinal axons and demyelination were manually excluded. Areas <0.09 µm2 were excluded from the analysis to eliminate the majority of small regions that did not correspond to axons. This methodology allowed an average sampling of 190 000 axons from 7–10 mice per group and includes almost all myelinated fibres (fibres with an axolemma diameter >0.3 µm). The data were plotted as relative frequency histograms per animal by dividing the number of axons with areas ranging from 0 to 80 µm2 (using 0.5 µm2 intervals as bins) by the total number of axons sampled. These frequencies were then averaged across the 7–10 animals per group. A graphic example of a relative frequency histogram for axons > 10 µm2 is shown for a group of uninfected SJL/J mice in Fig. 5A. Areas under the frequency histograms were also calculated to allow comparisons between the groups. Three different curve areas were determined to facilitate comparisons: (i) 0–4 µm2 (small axon fibres); (ii) 4–10 µm2 (medium to large axon fibres); and (iii) > 10 µm2 (largest axon fibres). These divisions were selected to aid in the graphic representation of where differences in area distributions were noted between groups. Statistical comparisons between curve areas were performed using a one-way ANOVA. Pairwise comparisons were done using the Student–Newman–Keuls method (P < 0.05). Correlation coefficients between large axon fibre frequencies and both demyelination percentages and rotarod performances were calculated using the Pearson product moment correlation (P < 0.05). (One mouse was excluded from the correlation in Fig. 6C because of poor tissue quality in the T6 section.) Data were also represented as ratios of infected divided by uninfected curve areas (Fig. 5B, C and D insets) for axons ranging from 0 to 80 µm2 (5 µm2 intervals for axonal areas ranging from 0 to 60 µm2; 20 µm2 intervals from 60 to 80 µm2) to illustrate the hierarchy of large axon fibre loss.
|
|
|
Electrophysiology
Motor-evoked potentials (MEPs) were elicited with stimulation of the spinal cord at the C8–T1 interspace and recorded in the hindlimb as previously described (Iuliano et al., 1994
; Rivera-Quiñones et al., 1998). Briefly, mice were anaesthetized by intraperitoneal injection of pentobarbital (0.08 mg/g). A rectal probe was used to monitor temperature which was maintained between 35.7 and 36°C. The stimulating electrodes were placed with the cathode inserted extradurally into the C8–T1 interspace and the anode subcutaneously 5–10 mm lateral to the cathode. The recording electrodes were placed in the hindlimb posterior to the middle third of the tibia (recording) and posterior to the calcaneus at the ankle (reference). A grounding electrode was placed subcutaneously over the upper lumbar spine. The stimulus intensity was calculated to yield a maximal response, usually obtained with a voltage of 8–10 V, and each recorded response was an average of 5–10 sweeps. Interelectrode distances were measured as previously described (Iuliano et al., 1994; Rivera-Quiñones et al., 1998). Conduction velocities at the onset of the MEPs were calculated by dividing inter-electrode distances by the respective latencies to onset. Amplitude was measured as the distance from baseline to the peak of the response. Each data point is the average of two recordings from one side of each mouse. Conduction velocities and latencies between the groups were compared statistically using a one-way ANOVA. Pairwise comparisons were made using the Student–Newman–Keuls method (P < 0.05). Amplitudes between 90 and 180 days post-injection (d.p.i.) were compared using an unpaired Student's t-test (P < 0.05).
Rotarod analysis
The Rotamex rotarod (Columbus Instruments; Columbus, OH, USA) was used to assess balance, coordination and motor control as described (McGavern et al., 1999b). The apparatus consists of a motor-driven rod suspended 28.5 cm over a grid and is capable of monitoring the performance of four mice at the same time through the use of lane dividers. Automation of the system is achieved through the use of a computer that records the velocity of the rod and the time at which a mouse falls and interrupts an infrared beam. Sham-infected (n = 7) and infected (n = 7) SJL/J mice were trained at 191 days after injection using a constant speed rotarod assay (speed = 5 r.p.m., time = 3 min, trials = 3) to familiarize them with the assay. This was followed by analysis of all mice at 192 days using an accelerated (7 r.p.m./min) rotarod assay (start speed = 5 r.p.m., end speed = 40 r.p.m., time = 5 min, trials = 3). The amount of time on the rotarod (seconds) was recorded for each of the three trials per mouse and then averaged. Statistical differences between sham-infected and infected SJL/J mice were calculated using a Student's t-test (P < 0.05). Correlations between rotarod performances and other morphological parameters were calculated using a Pearson product moment correlation P < 0.05).
|
Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
The percentage of spinal cord demyelination reaches a plateau at 100 days following TMEV infection
To examine the severity of spinal cord demyelination following TMEV infection, a time course experiment was conducted to quantify the percentage of demyelination in susceptible SJL/J mice infected for 45, 92–100 and 195–220 days (Fig. 1
). The time points were chosen to represent stages that delineate susceptibility versus resistance to virus-induced demyelination (45 d.p.i.) (Rodriguez and David, 1985), a decline in neurological function in susceptible SJL/J mice (92–100 d.p.i.) (McGavern et al., 1999b) and significant disruption of axon fibres in SJL/J mice (195–220 d.p.i.) (Rivera-Quiñones et al., 1998). We sectioned the spinal cord of infected SJL/J mice and quantified the total area of demyelination expressed as a percentage of the total white matter area. It has been observed previously that small, focal demyelinating lesions are present as early as 21 d.p.i. (Paya et al., 1990; McGavern et al., 1999b). This progresses to ~3% of the spinal cord white matter at 45 d.p.i. and 11% at 92–100 d.p.i. (Fig. 1A). Interestingly, the percentage of spinal cord white matter showing demyelination at 195–220 d.p.i. was not statistically different from that observed at 92–100 d.p.i. Therefore, the progressive demyelinating phase of the disease occurs during the first 100 days after infection and then reaches a plateau. Normal white matter is shown in Fig. 1B, whereas an example of the pathology observed in susceptible SJL/J mice is shown in Fig. 1C and D. Lesion size becomes progressively larger between 45 d.p.i (Fig. 1C) and 92–100 d.p.i. (Fig. 1D). This contrasts to resistant TMEV-infected C57BL/10 and uninfected SJL/J (Fig. 1B) mice that show no demyelinating lesions in the spinal cord white matter.
|
Prominent spinal cord atrophy occurs after the demyelinating phase of the disease
Spinal cord atrophy was assessed in susceptible SJL/J mice at 45, 92–100 and 195–220 d.p.i. Additionally, susceptible SJL/J mice were compared with TMEV-infected resistant C57BL/10 mice, which clear the virus within the first 21 days and do not show demyelination. These mice were studied at 195 d.p.i. to ensure that spinal cord atrophy was not the result of neuronal infection during the acute grey matter phase of the disease. Areas selected for analysis included: the total cord area, grey matter area, posterior white matter area and the remaining anterior and lateral white matter area. We have shown previously that only 3% of the demyelinating lesions were in the posterior columns of SJL/J mice infected for 180 days, whereas 30, 37 and 30% of the lesions were in the lateral, anterolateral and anterior columns, respectively (Rivera-Quiñones et al., 1998
). Thus, the posterior column area served as an important internal control for all groups, as we did not expect to observe morphological changes in this relatively preserved spinal cord region.
When total cord areas were compared between all groups (Fig. 2A), a significant reduction was only observed in SJL/J mice infected for 195–220 days. A 7% reduction was observed at the C1–C7 level of the spinal cord, and a 16% reduction was observed from C8–T11. To determine which anatomical region of the spinal cord was responsible for this reduction in total cord area, the remaining spinal cord areas were compared between the groups. As expected, no significant reduction was detected in the posterior white matter area or grey matter area for any of the groups (data not shown). In contrast, a significant reduction was observed in the anterior and lateral white matter area at the C1–C7 (12% reduction) and the C8–T11 (25% reduction) spinal cord levels for SJL/J mice infected for 195–220 days (Fig. 2B). A 13% area reduction was also observed at T12/13–L3; however, this was not statistically significant. At 45 and 92–100 d.p.i., a 12% reduction was detected at C8–T11 in SJL/J mice, yet this reduction was 2-fold less than that observed in SJL/J mice infected for 195–220 days (Fig. 2B). The more significant reductions detected at C8–T11 when compared with C1–C7 were most probably due to the high white to grey matter ratio in the thoracic spinal cord. Thus, pathological changes in the white matter were detected more easily. No reductions in cord area were observed in any cord regions analysed for resistant C57BL/10 mice infected for 195 days, suggesting that cord atrophy was the result of a chronic demyelinating disease in the susceptible strains and not the result of virus infection of neurons during the acute phase of the disease. Additionally, infected C57BL/10 cord sizes were comparable with uninfected SJL/J cord sizes, suggesting that the two strains had spinal cords of similar size.
|
Significant medium to large axon fibre loss contributes to the marked spinal cord atrophy
To examine area distributions of myelinated axon fibres in the non-demyelinated spinal cord, automated methodology was developed to sample a sizeable number of axons (15 000–30 000 axons) from the T6 section of each animal using the sampling scheme shown in Fig. 3E
. When relative frequency distributions from the lateral and anterior columns (Fig. 3E, fields B–H) were compared between uninfected, 45, 92–100 and 195–220 d.p.i. SJL/J mice, no statistically significant decreases were observed in the small myelinated axon fibres (0.1–4 µm2) for any of the groups (Fig. 3A). In contrast, significant medium to large myelinated fibre loss was detected in SJL/J mice infected for 195–220 days (Fig. 3B and C). A 30% reduction in medium to large fibres (4–10 µm2) was seen at 195–220 d.p.i., but not at 45 or 92–100 d.p.i. (Fig. 3B). In addition, a 78% reduction in large fibres (
10 µm2) was detected at 195–220 d.p.i., whereas only a 38% reduction was detected at 45 and 92–100 d.p.i. (Fig. 3C). As an internal control, axonal frequency distributions from the preserved posterior columns (Fig. 3E, field A) were also compared at each of the time points in SJL/J mice. As expected, there were no statistical differences between the frequencies of small fibres (97.3 ± 0.2, 98.1 ± 0.2, 96.8 ± 0.4 and 97.4 ± 0.5, respectively) or medium to large fibres (2.6 ± 0.3, 1.9 ± 0.2, 3.2 ± 0.4 and 2.6 ± 0.5, respectively) for uninfected, 45, 92–100 and 195–220 d.p.i. mice.
The pattern of large myelinated fibre (10 µm2) loss at 45, 92–100 and 195–220 d.p.i. was identical to the pattern of spinal cord atrophy observed in Fig. 2B. It is of note that the reduction in anterior and lateral column area was the same at 45 and 92–100 d.p.i. despite the significant increase in the severity of spinal cord demyelination at the latter time point. The large fibre loss (10 µm2) was also identical between 45 and 92–100 d.p.i. Interestingly, the most significant spinal cord atrophy was observed at 195–220 d.p.i., a time point when marked reductions were observed in two size categories of large myelinated axon fibres. However, there was no increase in the percentage of spinal cord demyelination from 92 to 100 d.p.i. This strongly supports the hypothesis that axon fibre loss results in severe spinal cord atrophy at the most chronic stage of the disease, after demyelination plateaus.
To determine if the severity of demyelination contributed to the loss of axons, the percentages of spinal cord demyelination were plotted against relative frequencies of medium to large myelinated axon fibres (4 µm2) for SJL/J mice infected for 195–220 days (Fig. 3D). The results revealed a strong negative correlation (r = –0.85, P = 0.008), suggesting that the animals with the most spinal cord demyelination had the greatest medium to large myelinated fibre loss. These results combined with the temporal profile for spinal cord myelin loss are consistent with the hypothesis that medium to large axon fibres are injured following demyelination.
Electrophysiological abnormalities result from a combination of demyelination and medium to large axon fibre loss
Electrophysiological abnormalities were assessed by eliciting and recording MEPs in the spinal cord (Iuliano et al., 1994; Rivera-Quiñones et al., 1998). We chose this method to assess global conduction abnormalities in CNS fibres objectively. Mice at 45, 90 and 180 d.p.i. were compared with uninfected SJL mice (Fig. 4). We reported previously a delayed conduction velocity at the peak of the response in SJL/J mice infected for 180 days (Rivera-Quiñones et al., 1998). This delay was concluded to be the result of the significant demyelination observed in the spinal cord white matter of these mice. To determine if the loss of medium to large myelinated fibres contributed to detectable electrophysiological abnormalities, three electrophysiological parameters (latency to onset, velocity to onset and amplitude) were selected for comparisons. Since it is known that the measurement of conduction velocity by this assay depends on the electrophysiological properties of a small percentage of the fastest conducting (or largest) axon fibres (McDonald and Sears, 1970), we hypothesized that as the largest myelinated fibres were lost, a population of smaller axons with slower conduction velocities would be responsible for the MEPs. This could contribute to a delayed latency and velocity to the onset of the response. We also assumed that large myelinated axon fibre loss would be manifest in a reduced amplitude of the response.
|
When the electrophysiological parameters were compared between the groups, a delayed latency to onset was observed at 45, 90 and 180 d.p.i. relative to uninfected SJL/J mice (Fig. 4A
). This delay was not statistically different between 45 and 90 d.p.i. (example in Fig. 4E); however, a more marked delay was noted at 180 d.p.i. (example in Fig. 4F). This trend was confirmed when the conduction velocities to onset were compared (Fig. 4B). Reductions in conduction velocities were detected at 45 and 90 d.p.i. that were not statistically different from one another, whereas the most significant reduction was noted at 180 d.p.i. Comparison of amplitudes supported the hypothesis that the significant loss of medium to large axon fibres at 180 d.p.i. contributed to electrophysiological abnormalities (Fig. 4C). The average amplitude for uninfected, 45 and 90 d.p.i. (examples in Fig. 4D and E) SJL mice was >17 µV, whereas the average for mice infected for 180 days (example in Fig. 4F) was 13 µV. A statistical comparison of amplitudes between 90 and 180 d.p.i. (two time points that differ significantly in the severity of large myelinated fibre loss) revealed a significant decrease at 180 d.p.i. (P < 0.02).
An illustration of the severity of medium to large myelinated fibre loss at 195–220 d.p.i. is shown in Figure 5. When relative frequency distributions were compared for axons >10 µm2 (Fig. 5), the most notable reduction in curve area was detected at 195–220 d.p.i. (Fig. 5D). In fact, the average area of fibres >10 µm2 was 18.8, 16.7, 16.0 and 14.8 µm2 for uninfected, 45, 92–100 and 195–220 d.p.i. SJL/J mice, respectively. Note that the averages were comparable at 45 and 92–100 d.p.i., whereas the most significant decrease was observed at 195–220 d.p.i. This concept is illustrated further in comparisons made between ratios of infected/uninfected axonal areas (Fig. 5B, C and D insets). A ratio of one signifies that there was no reduction in fibres when compared with uninfected SJL/J mice. A hierarchy in the loss of large myelinated fibres was evident and comparable at 45 d.p.i. (Fig. 5B inset) and 92–100 d.p.i. (Fig. 5C inset), whereas the small myelinated fibres (0–5 µm2) were completely preserved. A marked decrease in all categories of medium to large axon fibres (>5 µm2) was noted at 195–200 d.p.i. (Fig. 5D inset); however, the small fibres were preserved. These illustrations support the aforementioned electrophysiological results.
Neurological deficits correlate with spinal cord atrophy and medium to large myelinated fibre loss
To determine if spinal cord pathology correlated with the severity of neurological deficits in chronically infected SJL/J mice, a comprehensive study was conducted on 192-day-infected and sham (PBS)-injected SJL/J mice. Motor balance and coordination were assessed using the rotarod assay (Kuhn et al., 1995; Klugmann et al., 1997; Rozas and Labandeira Garcia, 1997; Griffiths et al., 1998; McGavern et al., 1999b). We have shown previously that disease progression can be monitored objectively with the rotarod assay in SJL/J mice infected with TMEV (McGavern et al., 1999b). The focus of this study was to correlate directly the functional impairment with spinal cord atrophy, medium to large myelinated axon loss and the percentage of spinal cord demyelination.
Following assessment of rotarod performance at 192 d.p.i., serial spinal cord sections were analysed from infected (n = 7) and sham-infected (n = 7) SJL/J mice to obtain a detailed representation of all cord regions. Assessment of cord atrophy in 192-day-infected versus sham-infected SJL/J mice confirmed previous findings in an independent experiment (Fig. 2). Analysis of total cord areas revealed an 11 and 20% reduction at C1–C7 and C8–T11, respectively. In addition, the combined lateral and anterior column areas were significantly reduced at C1–C7 (20%), C8–T11 (29%) and T12/13–L3 (15%); however, no reduction in the grey matter or posterior column area was observed. Calculation of myelinated axonal area distributions in the thoracic cord also revealed reductions in both medium and large axon size categories (17% reduction of fibres between 4 and 10 µm2; 62% reduction of fibres >10 µm2). Also in support of the previous results (Fig. 3), no reductions were observed in small myelinated axon fibres (0.1–4 µm2). Therefore, the quantitative morphological data regarding spinal cord areas and axonal dropout were confirmed in a second set of animals.
Comparison between sham-infected and infected SJL/J rotarod performances at 192 d.p.i. revealed a 67% reduction in the mean time on the rotarod for infected SJL/J mice. To determine the pathological parameters that best correlated with reduced rotarod performance for infected SJL/J mice, C7 combined lateral and anterior column areas (Fig. 6A), the percentages of spinal cord demyelination (Fig. 6B) and the frequencies of medium to large myelinated axon fibres (4 µm2) (Fig. 6C) were plotted against individual rotarod performances for infected SJL/J mice. Results revealed a near perfect positive correlation (r = 0.92, P = 0.008) between rotarod performances and the C7 combined lateral and anterior column area (Fig. 6A), demonstrating that animals with the most significant atrophy in this region had the greatest motor deficits. A weaker negative correlation (r = –0.66, P = 0.11) was obtained between rotarod performances and the percentage of spinal cord demyelination (Fig. 6B), suggesting that demyelination contributes to but is not the only parameter that causes functional impairment. In contrast, a near perfect positive correlation (r = 0.90, P = 0.013) was obtained between the frequency of medium to large myelinated axon fibres (4 µm2) assessed in the thoracic cord and rotarod performance (Fig. 6C). These data suggest that the C7 combined lateral and anterior column atrophy serves as an indicator of medium to large myelinated fibre loss. In fact, a strong positive correlation (r = 0.94, P = 0.02) was obtained between the C7 combined lateral and anterior column area and the frequency of medium to large myelinated axon fibres (4 µm2).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
This study demonstrates that medium to large myelinated axon loss (
4 µm2) is associated with electrophysiological abnormalities, and correlates with spinal cord atrophy and neurological deficits following demyelinating disease of the CNS. This conclusion is supported by the almost perfect correlation between large myelinated fibre loss, C7 combined lateral and anterior column atrophy and rotarod performance at the most chronic stage of the disease (190–220 d.p.i.). The most severe electrophysiological abnormalities were also observed at this chronic time point. These findings are in agreement with the MRI literature in multiple sclerosis, which states that NAA abnormalities in normal appearing white matter contribute significantly to disability (Davie et al., 1995, 1997; De Stefano et al., 1997, 1998; Naranyanan et al., 1997; Fu et al., 1998). The lack of a strong correlation between the percentage of spinal cord demyelination and rotarod performance at the most chronic stage of disease suggests that factors other than lesion load contribute to neurological deficits. This also supports the inability of MRI studies in multiple sclerosis to establish a strong association between T2 lesion load and EDSS score (Kidd et al., 1993, 1996; Filippi et al., 1995; Riahi et al., 1998). It is important to note that TMEV-induced demyelinating disease is restricted mainly to the spinal cord, whereas multiple sclerosis lesions are more widespread. This may contribute to the inability of MRI studies in multiple sclerosis to establish a correlation between lesion load and EDSS score. However, interestingly, the highest correlation (r = –0.67) described in a recent study evaluating the relationship between the EDSS and corticospinal tract lesion load (Riahi et al., 1998) was identical to the correlation obtained in this study (r = 0.66).
The temporal profile of spinal cord demyelination and medium to large myelinated fibre loss described in this study provides important insights into pathological correlates of clinical deficits. No changes in medium to large myelinated axon distributions or the degree of spinal cord atrophy were observed between 45 and 100 d.p.i. despite the significant increase (from 3 to 11%) in spinal cord demyelination. This suggests that the loss of axons observed at 45 and 100 d.p.i. is responsible for the observed spinal cord atrophy. If the atrophy was due to demyelination, the increase in demyelination from 3 to 11% (from 45 to 100 d.p.i.) should have been sufficient to induce further cord area reductions. Alternatively, further reductions in cord area at 100 d.p.i. may have been masked by compensatory oedema or gliosis due to increasing inflammation. In either case, comparable reductions in medium to large myelinated axon distributions were observed at 45 and 100 d.p.i. The findings that grey matter areas were preserved in SJL/J mice and that all spinal cord areas were preserved in TMEV-infected C57BL/10 mice suggest that the atrophy observed in SJL/J mice did not result from neuronal infection during the acute phase of the disease (within the first 21 d.p.i.). Between 100 and 200 d.p.i., a significant increase in spinal cord atrophy was observed without a concomitant increase in spinal cord demyelination, consistent with the hypothesis that atrophy resulted from axonal loss. In further support of this hypothesis, a near perfect correlation (r = 0.94) was observed between the C7 lateral/anterior column area and the frequency of medium to large myelinated fibres (4 µm2) at 192 d.p.i. These findings collectively suggest that axonal loss contributes to spinal cord atrophy during the course of demyelinating disease.
In a recent paper describing axonal transections in multiple sclerosis lesions, Trapp and colleagues concluded that a preferential attack of axons was unlikely, as most axons seemed to survive the demyelinating process and infrequently were associated with monocytic cells (Trapp et al., 1998). Our results also support this conclusion in demonstrating that axonal loss was not prominent until after the demyelinating phase of the disease was established. Demyelination increased until 100 d.p.i.; however, significant myelinated fibre loss was observed at 190–220 d.p.i., not 100 d.p.i. These findings were supported by the electrophysiological studies. It is likely that the early pattern of electrophysiological alterations (at 45 and 90 d.p.i.) is best explained by demyelination. The changes beyond this time point, when demyelination has reached a plateau, are best explained by the selective loss of the fastest conducting, large myelinated fibres. In concert, these results suggest that following the demyelinating process, naked axons are vulnerable to the inflammatory milieu and succumb to secondary damage. This could result in direct axonal injury or delayed Wallerian degeneration of myelin-deprived axons, which would explain the relatively long delay in the development of axonal loss. In either case, the secondary damage becomes manifest as a loss of normally myelinated axons. In fact, a strong negative correlation (r = –0.85) was observed between the percentage of spinal cord demyelination and the frequency of medium to large myelinated axon fibres (4 µm2).
It is also possible that decreases in the fibre diameters could contribute to shifts in the myelinated axonal frequency distributions. For example, atrophy to an area <4 µm2 would result in increased small myelinated fibre area frequencies and decreased large myelinated fibre frequencies. However, when the absolute number of small myelinated fibres (4 µm2) were compared in 192-day-infected and sham-infected SJL/J mice, only a 9% reduction was revealed that was not statistically significant. Additionally, no decreases in small fibre areas were detected. In contrast, 28 and 67% reductions were observed in the number of medium to large (4–10 µm2) and largest (10 µm2) fibres, respectively. These findings confirm that small myelinated fibres are relatively preserved and that axonal atrophy is not the primary factor contributing to decreased large fibre frequency distributions. It remains possible that some medium to large axon fibres shrink during the course of the demyelinating disease and replace small axon fibres that have died; however, the results from the electrophysiological studies are more consistent with a preferential loss of medium to large axon fibres. This preferential loss is also supported by the stepwise hierarchy of medium to large axon fibre loss shown in Fig. 5B, C and D (insets). In addition, previous studies of acute and chronic axonal loss following spinal cord contusion have also described selective losses of large calibre axons (Blight, 1983; Blight and Decrescito, 1986; Fehlings and Tator, 1995; Rosenberg and Wrathall, 1997).
This study has important implications in the field of demyelinating diseases, as it describes a temporal profile for spinal cord demyelination and axonal loss. The permanent disability resulting from axonal loss may be prevented if the factors that contribute to axonal damage are blocked following the demyelinating phase of the disease. Studies are under way to identify these factors and to design novel therapeutic interventions for the treatment of multiple sclerosis.
|
Acknowledgments |
|---|
This work was supported by the National Institutes of Health grants RO1 NS24180, RO1 NS32129, and the generous contributions of of Mr and Mrs Eugene Applebaum and Ms Kathryn Peterson. D.B.M. is supported by a predoctoral NRSA from the National Institute of Mental Health (Grant #1F31ME12120).
|
References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Adams RD, Victor M, Ropper AH. Multiple sclerosis and allied demyelinative diseases. In: Adams RD, Victor M, Ropper AH. Principles of neurology. 6th edn. New York: McGraw-Hill; 1997. p. 902–27.
Blight AR. Cellular morphology of chronic spinal cord injury in the cat: analysis of myelinated axons by line-sampling. Neuroscience 1983; 10: 521–43.[ISI][Medline]
Blight AR, Decrescito V. Morphometric analysis of experimental spinal cord injury in the cat: the relation of injury intensity to survival of myelinated axons. Neuroscience 1986; 19: 321–41.[ISI][Medline]
Charcot JM. Histologie de la sclerose en plaques. Gaz Hop (Paris) 1868; 41: 554–5, 557–8; 566.
Dal Canto MC, Lipton HL. Multiple sclerosis. Animal model: Theiler's virus infection in mice. Am J Pathol 1977; 88: 497–500.[ISI][Medline]
Dal Canto MC, Lipton HL. Recurrent demyelination in chronic central nervous system infection produced by Theiler's murine encephalomyelitis virus. J Neurol Sci 1979; 42: 391–405.[Medline]
Dal Canto MC, Lipton HL. Ultrastructural immunohistochemical localization of virus in acute and chronic demyelinating Theiler's virus infection. Am J Pathol 1982; 106: 20–9.[Abstract]
Davie CA, Barker GJ, Webb S, Tofts PS, Thompson AJ, Harding AE, et al. Persistent functional deficit in multiple sclerosis and autosomal dominant cerebellar ataxia is associated with axon loss. Brain 1995; 118: 1583–92.
Davie CA, Barker GJ, Thompson AJ, Tofts PS, McDonald WI, Miller DH. 1H magnetic resonance spectroscopy of chronic cerebral white matter lesions and normal appearing white matter in multiple sclerosis. J Neurol Neurosurg Psychiatry 1997; 63: 736–42.
De Stefano N, Matthews PM, Fu L, Narayanan S, Stanley J, Francis GS, et al. Axonal damage correlates with disability in patients with relapsing-remitting multiple sclerosis. Results of a longitudinal magnetic resonance spectroscopy study. Brain 1998; 121: 1469–77.
De Stefano N, Matthews PM, Narayanan S, Francis GS, Antel JP, Arnold DL. Axonal dysfunction and disability in a relapse of multiple sclerosis: longitudinal study of a patient. Neurology 1997; 49: 1138–41.
Fehlings MG, Tator CH. The relationships among the severity of spinal cord injury, residual neurological function, axon counts, and counts of retrogradely labeled neurons after experimental spinal cord injury. Exp Neurol 1995; 132: 220–8.[ISI][Medline]
Ferguson B, Matyszak MK, Esiri MM, Perry VH. Axonal damage in acute multiple sclerosis lesions. Brain 1997; 120: 393–9.
Filippi M, Paty DW, Kappos L, Barkhof F, Compston DA, Thompson AJ, et al. Correlations between changes in disability and T2-weighted brain MRI activity in multiple sclerosis: a follow-up study. Neurology 1995; 45: 255–60.[Abstract]
Filippi M, Campi A, Colombo B, Pereira C, Martinelli V, Baratti C, et al. A spinal cord MRI study of benign and secondary progressive multiple sclerosis. J Neurol 1996; 243: 502–5.[ISI][Medline]
Fu L, Matthews PM, De Stefano N, Worsley KJ, Narayanan S, Francis GS, et al. Imaging axonal damage of normal-appearing white matter in multiple sclerosis. Brain 1998; 121: 103–13.
Griffiths IR, Klugmann M, Anderson T, Yool D, Thomson S, Schwab MH, et al. Axonal swellings and degeneration in mice lacking the major proteolipid of myelin. Science 1998; 280: 1610–3.
Iuliano BA, Schmelzer JD, Thiemann RL, Low PA, Rodriguez M. Motor and somatosensory evoked potentials in mice infected with Theiler's murine encephalomyelitis virus. J Neurol Sci 1994; 123: 186–94.[ISI][Medline]
Kidd D, Thorpe JW, Thompson AJ, Kendall BE, Moseley IF, MacManus DG, et al. Spinal cord MRI using multi-array coils and fast spin echo. II. Findings in multiple sclerosis. Neurology 1993; 43: 2632–7.
Kidd D, Thorpe JW, Kendall BE, Barker GJ, Miller DH, McDonald WI, et al. MRI dynamics of brain and spinal cord in progressive multiple sclerosis. J Neurol Neurosurg Psychiatry 1996; 60: 15–9.
Klugmann M, Schwab MH, Puhlhofer A, Schneider A, Zimmermann F, Griffiths IR, et al. Assembly of CNS myelin in the absence of proteolipid protein. Neuron 1997; 18: 59–70.[ISI][Medline]
Kuhn PL, Petroulakis E, Zazanis GA, McKinnon RD. Motor function analysis of myelin mutant mice using a rotarod. Int J Dev Neurosci 1995; 13: 715–22.[ISI][Medline]
Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status score (EDSS). Neurology 1983; 33: 1444–52.
Lehrich JR, Arnason BG, Hochberg FH. Demyelinative myelopathy in mice induced by the DA virus. J Neurol Sci 1976; 29: 149–60.[ISI][Medline]
Lipton HL. Theiler's virus infection in mice: an unusual biphasic disease process leading to demyelination. Infect Immun 1975; 11: 1147–55.
Lipton HL, Dal Canto MC. Chronic neurologic disease in Theiler's virus infection of SJL/J mice. J Neurol Sci 1976a; 30: 201–7.[ISI][Medline]
Lipton HL, Dal Canto MC. Theiler's virus-induced demyelination: prevention by immunosuppression. Science 1976b; 192: 62–4.
Losseff NA, Webb SL, O'Riordan JI, Page R, Wang L, Barker GJ, et al. Spinal cord atrophy and disability in multiple sclerosis. Brain 1996; 119: 701–8.
Lycklama a Nijeholt GJ, van Walderveen MA, Castelijns JA, van Waesberghe JH, Polman C, Scheltens P, et al. Brain and spinal cord abnormalities in multiple sclerosis. Correlation between MRI parameters, clinical subtypes and symptoms. Brain 1998; 121: 687–97.
McDonald WI, Sears TA. The effects of experimental demyelination on conduction in the central nervous system. Brain 1970; 93: 583–98.
McGavern DB, Murray PD, Rodriguez M. Quantitation of spinal cord demyelination, remyelination, atrophy, and axonal loss in a model of progressive neurologic injury. J Neurosci Res 1999a. In press.
McGavern DB, Zoecklein L, Drescher KM, Rodriguez M. Quantitative assessment of neurologic deficits in a chronic progressive murine model of CNS demyelination. Exp Neurol 1999b; 158: 171–81.[ISI][Medline]
Narayanan S, Fu L, Pioro E, De Stefano N, Collins DL, Francis GS, et al. Imaging of axonal damage in multiple sclerosis: spatial distribution of magnetic resonance imaging lesions. Ann Neurol 1997; 41: 385–91.[ISI][Medline]
Oppenheimer DR. The cervical cord in multiple sclerosis. Neuropathol Appl Neurobiol 1978; 4: 151–62.[ISI][Medline]
Paya CV, Leibson PJ, Patick AK, Rodriguez M. Inhibition of Theiler's virus-induced demyelination in vivo by tumor necrosis factor alpha. Int Immunol 1990; 2: 909–13.
Riahi F, Zijdenbos A, Narayanan S, Arnold D, Francis G, Antel J, et al. Improved correlation between scores on the expanded disability status scale and cerebral lesion load in relapsing-remitting multiple sclerosis. Brain 1998; 121: 1305–12.
Rivera-Quiñones C, McGavern D, Schmelzer JD, Hunter SF, Low PA, Rodriguez M. Absence of neurological deficits following extensive demyelination in a class I-deficient murine model of multiple sclerosis. Nature Med 1998; 4: 187–93.[ISI][Medline]
Rodriguez M, David CS. Demyelination induced by Theiler's virus: influence of the H-2 haplotype. J Immunol 1985; 135: 2145–8.[Abstract]
Rodriguez M, Leibowitz JL, Lampert PW. Persistent infection of oligodendrocytes in Theiler's virus-induced encephalomyelitis. Ann Neurol 1983; 13: 426–33.[ISI][Medline]
Rodriguez M, Oleszak E, Leibowitz J. Theiler's murine encephalomyelitis: a model of demyelination and persistence of virus. [Review]. Crit Rev Immunol 1987; 7: 325–65.[Medline]
Rosenberg LJ, Wrathall JR. Quantitative analysis of acute axonal pathology in experimental spinal cord contusion. J Neurotrauma 1997; 14: 823–38.[ISI][Medline]
Rozas G, Labandeira Garcia JL. Drug-free evaluation of rat models of parkinsonism and nigral grafts using a new automated rotarod test. Brain Res 1997; 749: 188–99.[ISI][Medline]
Stevenson VL, Leary SM, Losseff NA, Parker GJ, Barker GJ, Husmani, Y, et al. Spinal cord atrophy and disability in MS. A longitudinal study. Neurology 1998; 51: 234–8.
Trapp BD, Peterson J, Ransohoff RM, Rudick R, Mork S, Bo L. Axonal transection in the lesions of multiple sclerosis. N Engl J Med 1998; 338: 278–85.
Urenjak J, Williams SR, Gadian DG, Noble M. Proton nuclear magnetic resonance spectroscopy unambiguously identifies different neural cell types. J Neurosci 1993; 13: 981–9.[Abstract]
Received May 18, 1999. Revised September 3, 1999. Accepted September 28, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|