1. Introduction
Voltage-gated sodium channels (NaChs) are highly expressed within those regions of excitable tissues that play a critical role in the initiation and propagation of action potentials (Hille, 2001). One such region is the node of Ranvier where NaCh activation is necessary for saltatory conduction. The node of Ranvier also serves as a model to study the targeting of NaChs to the axolemma (Poliak and Peles, 2003; Salzer and James, 2003). Recently, there has been interest in the role of altered NaCh expression to the generation of pain states (Lai et al., 2003, 2004; Wood, 2004). Clinical experience also shows that many of the commonly used medications that show effectiveness in the treatment of neuropathic pain conditions have a NaCh blocking effect (Jensen, 2002; McQuay, 2002). In addition, NaChs are a diverse group consisting of at least nine different isoforms that are expressed not only differentially within the normal nervous system (Goldin et al., 2000) but that also show unique changes in expression associated with persistent pain states (Wood et al., 2004). These changes in isoform expression have been implicated with the development of increased neuronal excitability (Suzuki and Dickenson, 2000) leading to the activation of pain pathways and thus represent attractive pharmacological targets for the development of new analgesics. The isoforms that appear as the most attractive targets include those that are preferentially expressed in the peripheral nervous system (so to avoid CNS side-effects) and that increase their expression after nerve injury or with inflammation.
There is evidence for a role of altered NaCh expression contributing to both inflammatory (Black et al., 2004) and neuropathic (Lai et al., 2003) pain conditions. As more is learned about these different conditions, the distinctions between the two become less (DeLeo and Yezierski, 2001; Watkins and Maier, 2002) and because of these similarities it is possible that a new analgesic may be effective in both conditions. Here we utilize a nerve lesion model that has components of both inflammation, as induced by chromic gut suture (Clatworthy et al., 1995), and a partial nerve axotomy/neuropathic lesion that leaves some fibers spared. This model is a modification of an earlier one that exposes the ION at the medial aspect of the orbit (Gregg, 1973) and then chromic sutures applied (Vos et al., 1994), but with the addition of a partial axotomy. We have used a less invasive approach that involves the exposure and manipulation of the nerve distal to the IO foramen.
Even though some NaCh isoforms have been especially implicated with pain conditions, a rigorous quantitative evaluation of the expression of the various isoforms and overall NaChs in fibers at the site of the lesion has not been performed. In this study we present a method that allows a quantitative analysis of NaCh expression within accumulations including those located at nodes of Ranvier and demonstrate a striking alteration in NaCh distributions and immunofluorescence intensity following an ION lesion that produces behavior showing increased sensitivity to mechanical stimuli.
2. Methods
2.1. Animals and surgical procedures
This study was approved by the University of Colorado Health Sciences Center Animal Care and Use Committee. Five young adult female Sprague–Dawley rats were used as experimental subjects, while four were used as normal control subjects in this study. The experimental subjects were anesthetized with an intraperitoneal injection of pentobarbital (10mg per 200–225g subject) and once pain-free the left ION was exposed just distal to the infraorbital foramen by way of a midline incision over the snout. The connective tissue overlying the ION was carefully removed until the multiple fascicles that comprise the nerve were visualized. The ION was released from the deeper tissues by passing a blunt probe under the nerve. Two 4-0 chromic gut sutures (Ethicon Inc., Somerville, New Jersey) were then placed around the ION just distal to the foramen, and the nerve was partially constricted as the sutures were tightened. The lateral half of the ligated fibers were then transected just distal to the sutures, while the medial half of the fibers were left intact and thus spared. The superficial incision was closed with Dexon 4-0 suture. Experimental subjects were given acetaminophen in drinking water (2mg/ml) two days prior to surgery that continued through the fourth day after surgery.
2.2. Behavioral testing
Following a two-week survival period experimental subjects were placed into a clear plastic box (30cm wide, 20cm deep, 25cm high) and then tested for a threshold-withdrawal response to graded stimulation of the vibrissae pads with Semmes–Weinstein monofilaments (“Touch-Test Sensory Evaluator”; North Coast Medical Inc., Morgan Hill, CA). Three of the four control subjects were also evaluated for threshold-withdrawal responses in a manner similar to the experimental subjects. Stimulation was performed with 1, 2, 4, 6, 8, 10, 15 and 26g monofilaments in alert subjects without administration of any pharmacologic agents. The center of each vibrissae pad was contacted at a 90 degree angle with the tip of the filament and force applied until filament bending occurred and then the behavior noted. The right and then left vibrissae pads were successively stimulated with monofilaments of ascending size until a threshold response was noted on each side. Threshold was defined as a response of “2” or greater according to the following scale; 0=no response or grabbing of the filament after stimulus, 1=swipe at face one time after stimulus, 2=withdraw and/or bite at filament after stimulus, 3=attack filament after stimulus, 4=prolonged directed facial grooming and/or scratching after stimulus. Five minutes were allowed between each stimulus application.
2.3. Tissue processing and staining
After sensory testing, experimental and control subjects were again anesthetized with an intraperitoneal injection of pentobarbital (25mg) and then transcardially perfused with 100ml of 0.9% saline in H2O followed by 250ml of fixative consisting of 4% paraformaldehyde in 0.1M phosphate buffer (PB) at pH 7.4. The ION was exposed and removed extending from its origin at the maxillary/ophthalmic division root to its termination in the vibrissa pad and then post-fixed in the same fixative for 20min at room temperature. The sutures were left in place around the nerves of experimental subjects for orientation purposes. Fixed nerves were rinsed three times in 0.1M PB for 10min each, followed by a rinse in 0.1M PB with 15% sucrose at room temperature for one hour. The IONs were then placed in 0.1M PB with 30% sucrose overnight on a shaker at 4°C.
Separate IONs from different subjects were placed side-by-side, embedded in mounting medium (Neg-50; Richard-Allan Scientific, Kalamazoo, MI) and sectioned in the horizontal plane at 30–35μm with the use of a cryostat. Sections were placed onto Superfrost Plus slides (Fisher Scientific, Pittsburgh, PA), allowed to dry and stored in a −20°C freezer. The slides with specimens were removed from the freezer prior to staining and all subsequent steps described below were performed at room temperature. The specimens were rinsed three times in 0.1M phosphate buffered saline (PBS) for 10min in each rinse. Non-specific binding was decreased by the incubation of the tissue with 2% bovine γ-globulin (Sigma, St. Louis, MO), 4% normal goat serum (Sigma) and 0.3% Triton X-100 (Fisher Scientific) in 0.1M PBS for 90min as a blocking solution. The blocking solution was removed and the tissue was incubated in primary antibodies diluted in blocking solution and placed overnight in a humidifier.
A “pan-specific” NaCh antibody that is based on an epitope of the α-subunit conserved in all vertebrate NaCh isoforms (Dugandzija-Novakovic et al., 1995; Kaplan et al., 1997) was used at a 1:100 concentration, This antibody recognizes a single band, at the appropriate molecular weight (∼230kDa), on Western blots of separation gels made with rat brain derived membrane proteins (Kaplan et al., 1997) and the brain tissue labeling on Westerns and nodes of Ranvier in immunofluorescence is eliminated by preincubation of the antibody with peptide against which the antibody is raised (Dugandzija-Novakovic et al., 1995; Kaplan et al., 1997). A caspr (paranodin) monoclonal antibody (Poliak et al., 1999) was used at a 1:500 concentration to identify nodes of Ranvier that are seen as gaps between the paranodal staining of caspr. The next day the tissue was rinsed in 0.1M PBS followed by incubation in secondary antibodies diluted to a 1:100 concentration in blocking solution for 90min in a humidifier while protected from the light. An Alexa Fluor 568-conjugated anti-rabbit IgG secondary antibody (red fluorophore; Molecular Probes, Eugene, OR) was used to visualize the NaCh-immunoreactivity (IR), while an Alexa Fluor 488-conjugated anti-mouse IgG secondary antibody (green fluorophore; Molecular Probes) was used to visualize the caspr-IR. Tissues were rinsed in 0.1M PBS, then water, allowed to dry, coverslipped with Vectashield (Vector Labs, Burlingame, CA) and stored at 4°C. Control tissue specimens were processed as above except the undiluted NaCh antibody was preincubated with peptide antigen (approx. 30:1 peptide to antibody molar concentration ratio) for a minimum of four hours before the peptide-blocked antibody was diluted and applied to the tissue. Tissue specimens were evaluated with a Nikon PCM-2000 laser scanning confocal microscope. The association of NaCh-IR with caspr-IR was quantitatively evaluated as described below. SimplePCI software, v4.0 (Compix Inc., Cranberry Township, PA) was used for acquisition of all images. Final image processing for illustration purposes was done with Adobe Photoshop CS or Corel Photo-Paint 12.
2.4. Quantification of NaCh staining
Longitudinal sections of the IONs that had been double-stained with NaCh and caspr antibodies as described above were examined with the confocal microscope. A “z-series” was obtained through each section, at 0.8μm steps, using a 40X/N.A. 1.30 oil immersion objective. Laser gain levels were the same for all acquired images and were selected at levels that allowed optimal visualization of the fluorophores used as secondary antibodies. The gain level used to visualize the NaCh staining was selected at a level that typically allowed the identification of multiple pixels with a maximum 255 intensity (see below) in most nodes of Ranvier, while still allowing an obvious gradation of staining intensities within these NaCh accumulations. The tissues were minimally evaluated prior to capture of the z-series so as to avoid photo-bleaching. Z-series images were first obtained of NaCh-IR followed by a two laser scan that captured NaCh-IR as red and caspr-IR as green. Similar regions of the nerve in normal and lesioned IONs were selected for imaging and corresponded to the region approximately 0.2mm proximal to the placement of the chromic sutures in the experimental subjects. Each image was saved at a resolution of 1024×1024 pixels, with each square pixel being 0.3μm in length and representing an area of 0.09μm2.
Using ImageJ software (from NIH and available at http://rsb.info.nih.gov/ij/), both corresponding z-series stacks (NaCh alone, and the two color NaCh and caspr) were then opened simultaneously (Fig. 1), with each stack displayed on its own monitor in a multi-screen array. The NaCh accumulations were then evaluated for their relationship or lack of relationship with caspr in the two color image while navigating through the various levels of each z-series. Only those NaCh accumulations that were totally contained within the z-dimension of the stack were evaluated. The NaChs that were adjacent to caspr-positive clusters were classified as either; (1) typical nodes – NaCh fills the gap at the node of Ranvier as identified by the paranodal staining of caspr on both sides of the node, (2) split nodes – two distinct NaCh accumulations, separated by a gap in the NaCh staining within the same fiber and with each NaCh accumulation flanked on one side with caspr staining, or (3) heminodes – caspr staining located on only one side of a contiguous NaCh accumulation, while those NaCh accumulations that lacked an association with caspr were classified as “naked” accumulations (Fig. 1A).
;)
Fig. 1: Illustration of the methods used to classify and measure sodium channel (NaCh) accumulations. (A) A two-color confocal micrograph of NaCh (red) and caspr (green) staining obtained from a lesioned infraorbital nerve that represents one level of a z-series. The NaCh accumulations are prominent at caspr-related sites that include typical nodes (white arrow), split nodes (black arrow), and heminodes (white arrowhead), while accumulations that lack caspr are seen as “naked” accumulations (black arrowhead). (B) Demonstrates the relationship of a caspr-associated NaCh accumulation at different levels of the z-series within a typical node (bottom box in A) while (C) demonstrates a “naked” NaCh accumulation that lacks an association with caspr (top box in A). The distance between each image in the z-series is 0.8 μm. The NaCh accumulation of interest is examined in the corresponding immunofluorescence z-series (B′ and C′ grey-level series) and the level showing the greatest number of contiguous pixels (*; red box) with maximum 255 intensity is calculated as described in the text. The fiber diameter of caspr-associated accumulations is determined by measuring the maximum width of caspr staining in the slice with the most prominent staining (white line in B – 4th image from left). Scale bar in A = 10 μm.
Once the NaCh accumulation was classified based on its relationship with caspr in the two color image (Fig. 1B and C), the same NaCh accumulation was visually examined in various levels of the NaCh immunostained (black and white) z-series to select the image/level that showed maximum size and intensity of staining (Fig. 1B′ and C′). The area containing the maximum NaCh accumulation in the NaCh only stack was then chosen as a region of interest (ROI) with the “wand (tracing) tool” and this function was used to determine the number of pixels within this accumulation that showed a maximum pixel intensity of 255 (on a 0–255 scale; red boxes in panels with * in Fig. 1B′ and C′). The actual numbers of pixels were obtained with the histogram function found under the “Process” menu. All caspr-associated NaCh accumulations were evaluated for the presence of maximum intensity NaCh pixels, while the analysis of “naked” NaCh accumulations was limited to those accumulations that consisted of at least four or greater contiguous pixels. In the rare instances (44/1180) where there were other maximum intensity pixels that were obviously associated with, but discontiguous from the larger accumulation, the total numbers of all maximum intensity pixels were summed. The paranodal fiber diameter associated with each caspr-associated accumulation was determined by measuring the maximum width of caspr staining within the paranodal region(s) (Fig. 1B; 4th panel from the left). Note that the image section used for this measurement was at times different from that used to measure maximum NaCh area (Fig. 1B and B′); this is expected because the caspr staining would appear widest in optical sections containing the axial center of the nodal axon, while the maximum NaCh area would be expected in sections maximally encompassing the top or bottom surface of the nodal axolemma.
Once measured, a marker was placed over each NaCh accumulation in every image in the stack that it appeared to avoid multiple evaluations of the same accumulation, and the results from the analysis of each accumulation were recorded in an Excel worksheet. All evaluations were done by one investigator to minimize variability in assessment.
Since the NaCh accumulations appeared to vary so dramatically after the ION lesion, an additional ImageJ analysis was performed on evenly spaced images within each z-series to determine the density and average size of NaCh accumulations within a given area of the IONs. At least three and up to six different images within each NaCh z-series that were separated by 5.6μm were used for this purpose. This distance was used to avoid multiple counts of the same accumulations. In ImageJ, the area containing nerve tissue within each image was outlined with the “freehand selections” tool and added to the ”ROI Manager” function found under the “Analyze” menu. This was done to exclude areas devoid of fibers. The “Measure” function in the “ROI Manager” was used to determine the total area of the ROI. The image was thresholded to limit the analysis to pixels with a maximum intensity of 255. The “Analyze Particles” function under the “Analyze” menu was employed to determine the total count and average size of accumulations that had four or more pixels of maximum intensity within the ROI.
Statistical analyses to determine the significance of differences included the use of a Student paired t-test for behavioral data, two-way ANOVA for Fig. 5 analysis, and unpaired Student t-test for all other data. Error bars on all graphs represent the standard error of the mean.
Fig. 5: The proportions of atypical sodium channel (NaCh) clusters increase in lesioned infraorbital nerves (IONs). Note the dramatic shift away from typical nodes to altered forms (heminodes, split nodes and naked accumulations) in lesioned IONs compared with normal IONs. For statistical analysis, see the text.
3. Results
3.1. Behavioral response to monofilament stimulation of vibrissa pads shows increased sensitivity to mechanical stimuli on the side of the lesion in experimental subjects
Monofilament stimulation of the vibrissae pads two weeks after the left ION lesion showed that significantly (p≤0.001) less force was needed to produce a behavioral response of “2” or greater following stimulation of the left vibrissa pad when compared to the contralateral (right) vibrissa pad in experimental subjects (Fig. 2). An exaggerated behavioral response (“4” response in Fig. 2) to stimulation of the vibrissa pad on the side of the lesion was seen in two of these subjects and consisted of prolonged facial grooming directed to the side of the lesion. The largest monofilament was limited to 26g since larger ones delivered so much force they resulted in lateral displacement of the head that complicated detection of the withdrawal response. Behavioral response to mechanical stimulation of the vibrissae pad was evaluated in three of the normal subjects and this evaluation showed the lack of a threshold response after stimulation of either vibrissa pad with all stimuli applied (Fig. 2). When there was no threshold response of “2” or greater following the 26g monofilament stimulation the test was terminated and this result was indicated by an * as seen in Fig. 2.
Fig. 2: Infraorbital nerve (ION) lesions increase behavioral sensitivity to touch stimuli. The graph shows the gram force needed to obtain a threshold response of “2” or greater following monofilament stimulation of the right and left vibrissal pads. Non-threshold responses include; 0 = no response or grab at filament, and 1 = swipe at face once after stimulation. Threshold responses include; 2 = withdraw and/or bite at filament, 3 = attack filament, and 4 = prolonged directed grooming and/or scratching of the stimulated vibrissa pad. Monofilaments (1, 2, 4, 6, 8, 10, 15 and 26 g) are applied to the right and left vibrissal pads in ascending order until a threshold response occurs and the number associated with each bar indicates the type of threshold response elicited. Asterisks (*) indicate groups where no threshold response is achieved. The force needed to elicit a threshold response after stimulation of the left vibrissa pad in ION lesioned subjects was significantly less (p ≤ 0.001) when compared to the non-lesioned right side.
3.2. Confocal microscopy reveals altered NaCh distributions in lesioned IONs
The confocal microscopic evaluation of NaCh-immunoreactivity (IR) demonstrated a dramatic overall change in the distribution, immunofluorescence intensity and morphology of NaCh accumulations at nodal sites identified by caspr staining and the presence of dense accumulations at caspr-negative sites in lesioned IONs when compared to that seen in control subjects (Fig. 3). While examination of IONs from normal subjects showed mostly normal nodes (Fig. 3A and B), the NaCh accumulations seen in lesioned IONs appeared more prominent and included ones with altered caspr relationships (Fig. 3C and D). The shape of NaCh accumulations in the lesioned IONs appeared more round (Fig. 3D) and thus different from the disc-like appearance usually seen at nodes in normal IONs (Figs. 3B and 4A). The altered forms of NaCh accumulations seen in lesioned IONs included typical nodes with an increased length of NaChs in the nodal gap that made the node appear “robust” (Fig. 4B), split nodes (Fig. 4C and D), heminodes (Fig. 4E), and the presence of “naked” accumulations that lacked an association with caspr (Fig. 4F). Since the most prominent difference in the localization of NaChs seen in the lesioned IONs was the presence of large accumulations of brightly stained NaCh clusters at normal and altered caspr and non-caspr related sites, these accumulations were further evaluated in a quantitative fashion.
Fig. 3: Representative confocal micrographs showing the density and size of sodium channel (NaCh) accumulations seen within longitudinally sectioned infraorbital nerves (IONs) from normal (A and B) and lesioned (C and D) subjects. The micrographs in the top row (A and C) show the combined NaCh (red) and caspr (green) image, while the corresponding grey-level NaCh image (B and D) is seen immediately below. (A and B) The prominent NaCh staining seen within normal IONs (A and B) is located at nodes of Ranvier (arrow in A) with intact paranodal staining of caspr (arrowhead in A). The high density of NaChs at the nodes appears disc-like (arrow in B). (C) The NaCh staining seen in lesioned IONs is prominent at typical nodes (arrow) and at sites with altered caspr relationships (arrowheads). (D) The NaCh accumulations seen in lesioned IONs appear more numerous and larger (arrows) than those seen in B. Scale bar in A = 10 μm and also applies to B–D.
Fig. 4: Confocal micrographs showing a typical node from a normal infraorbital nerve (ION) (A) and the different types of caspr-related sodium channel (NaCh) accumulations commonly seen after an ION lesion (B–F). The NaCh accumulations (red; arrowheads) are either directly adjacent to caspr clusters (green) (A–E) or lack an association with caspr (F). (A) The NaChs are confined to a narrow region in a typical node from a normal ION. (B–F) The NaCh accumulations seen in lesioned IONs consist of caspr-associated forms that include typical nodes that appear elongated or “robust” (B), split nodes (C and D), and heminodes (E), while some NaCh accumulations lack caspr and thus appear “naked” (F).
3.3. Quantitative analysis shows a shift from typical nodes to altered forms indicative of demyelination in lesioned IONs
The relative proportions of different types of NaCh accumulations were determined in lesioned and normal IONs (Fig. 5). The accumulations were classified as those types found in association with caspr (typical node, split node or heminode) and those that lacked caspr and thus classified as “naked” accumulations. The total numbers of accumulations analyzed were 1180, with 535 in normal IONs, and 645 in the lesioned IONs. Most of these accumulations were associated with caspr (518/535 in normal IONs and 533/645 in lesioned IONs). The vast majority of caspr-related accumulations seen in normal IONs consisted of typical nodes (93.8%), while split nodes (1.3%) and heminodes (1.7%) were both very rare. The analysis of caspr-related NaCh accumulations in lesioned IONs showed a dramatic shift away from the high proportion of typical nodes seen in the normal IONs, with an increased prevalence of both split nodes and especially heminodes. The caspr-associated NaCh accumulations in lesioned IONs consisted of 51.5% typical nodes, 25.0% heminodes, and 6.2% split. The proportion of “naked” NaCh accumulations relative to the total number of all NaCh accumulations evaluated also dramatically increased in the lesioned IONs (112/645; 17.4%) when compared to those seen in the normal IONs (17/535; 3.2%). A two-way ANOVA test showed that the difference in the frequency of all caspr-related and “naked” accumulations in the lesioned IONs was significantly different than those frequencies of each type seen in normal IONs (p<0.001 in all cases).
Overall these findings demonstrate that the relative proportion of typical nodes decreases in lesioned subjects with a concurrent increase in split nodes, and especially heminodes and “naked” accumulations. The shift from NaCh accumulations associated with typical nodes to other caspr-related sites most likely results from significant demyelination secondary to the nerve lesion, since both split nodes and heminodes are associated with experimentally induced demyelination (Dugandzija-Novakovic et al., 1995; Novakovic et al., 1996; Arroyo et al., 2004).
3.4. Quantitative analysis shows increased immunofluorescence intensity in individual NaCh accumulations in lesioned IONs
In an effort to determine if there was a change in the expression of NaChs within accumulations in lesioned IONs, an analysis was performed on the number of pixels that showed maximum NaCh immunofluorescence intensity (see Methods) within typical nodes. The analysis was limited to typical nodes since these were the only caspr-related sites that were common among both normal and lesioned IONs. This analysis allowed a determination of the total area of pixels with maximum intensity since each pixel represents an area of 0.09μm2. The result of this analysis showed that typical nodes in the lesioned IONs contained about three times the number of maximum intensity NaCh pixels when compared to those pixels in typical nodes of normal IONs (Fig. 6; 48.8 pixels in lesioned, vs. 16.8 pixels in normal IONs). This increased area of NaCh pixels with maximum immunofluorescence intensity seen in lesioned IONs was highly significantly different (p≤0.0001) than those seen in normal IONs.
Fig. 6: The sodium channel (NaCh) immunofluorescence intensity of nodes themselves increases in lesioned infraorbital nerves (IONs). A comparison of the mean number of NaCh pixels with maximum intensity measured in typical nodes of normal and lesioned IONs shows that typical nodes in lesioned IONs contain significantly more maximum intensity pixels than those seen in normal IONs (* p ≤ 0.0001).
It is possible that the increased area of maximum intensity pixels simply reflected a commensurate increase in nodal area in lesioned IONs, e.g. as seen in the phenomenon of “robust” nodes. However, there are other indications that the NaCh nodal surface density also increased. Thus, although all nodes examined contained bright NaCh staining, there were some typical nodes that did not contain any pixels with maximum intensity at the laser gain levels used; these included 67/502 in normal IONs, whereas none of the 332 typical nodes evaluated in the lesioned IONs lacked maximum intensity pixels. This finding strongly supports an augmentation of NaChs at intact nodes as an axonal response to injury.
Since a measurement of paranodal diameter was performed on the caspr staining associated with the NaChs at typical nodes, scatter plots were generated that showed the relationship between the number of pixels with maximum NaCh intensity as a function of paranodal diameter size in lesioned and normal IONs (Fig. 7). The fitted regression line in the scatter plot for typical nodes in normal IONs (Fig. 7A) suggests that the number of maximum intensity pixels is independent of paranodal fiber diameter. However, the nodes of lesioned IONs demonstrated an increase in the numbers of pixels with maximum NaCh intensity. Further, this increase appeared to be positively correlated with fiber diameter, suggesting that the underlying mechanisms of nodal NaCh augmentation were increased proportionally as nodal size increased (Fig. 7B).
Fig. 7: Sodium channel (NaCh) nodal immunofluorescence intensity increases in all fibers in lesioned infraorbital nerves (IONs). The number of maximum intensity NaCh pixels seen in typical nodes is scatter-plotted as a function of paranodal diameter in normal (A) and lesioned (B) IONs and the data fitted by linear regression. Comparison of the linear regression line parameters suggests that in normal IONs the number of maximum intensity pixels was relatively constant and independent of paranodal fiber diameter (i.e. the slopes of the regression lines were not significantly different from zero). In contrast, the regression slope was highly significantly non-zero and positive (p ≤ 0.0001) in lesioned IONs, thus suggesting an increase in NaCh expression that is proportional to paranodal fiber diameter.
An evaluation of the number of maximum intensity NaCh pixels within the “naked” accumulations (rare in normal IONs but common in lesioned IONs) was also performed (Fig. 8). This evaluation also showed a significant (p≤0.0004) increase in maximum intensity NaCh pixels within “naked” accumulations seen in the lesioned IONs (an average of 42.6 pixels) when compared to normal IONs (an average of 16.4 pixels). By this measure, there was a more than twofold increase of maximum intensity immunofluorescence NaCh pixels in “naked” accumulations in lesioned IONs when compared to normal IONs. Interestingly, the number of maximum intensity NaCh pixels seen in “naked” accumulations in lesioned IONs (42.6 pixels) also approached the number seen in typical nodes of lesioned IONs (48.8 pixels).
Fig. 8: “Naked” sodium channel accumulations in lesioned infraorbital nerves (IONs) also contain significantly more maximum intensity pixels than those seen in normal IONs (* p ≤ 0.0004).
3.5. Quantitative analysis shows an increased number and size of NaCh accumulations in lesioned IONs
An analysis of the number and average size of all forms of NaCh accumulations consisting of four or more contiguous pixels of maximum intensity was performed among the lesioned and normal IONs. The number of accumulations expressed as a function of nerve area revealed a significant difference (p≤0.0001) in this accumulation density in lesioned IONs when compared to normal IONs (Fig. 9). The difference in NaCh accumulation density was greater than a threefold increase within the lesioned IONs (449 accumulations/mm2) when compared to normal IONs (143 accumulations/mm2). The average size analysis of these accumulations showed that the accumulations in lesioned IONs (28.0 pixels) were also significantly (p≤0.0001) larger than those seen in the normal IONs (16.1 pixels; Fig. 10). These calculations were based on 334 accumulations in normal IONs and 737 accumulations in lesioned IONs. Overall these results show a dynamic reorganization and augmentation of NaCh density within lesioned IONs that includes not only an increased size of individual NaCh accumulations, but also an increased overall number of such clusters.
Fig. 9: The total field density of sodium channel (NaCh) accumulations increases dramatically in lesioned infraorbital nerves (IONs). The NaCh accumulations with maximum pixel intensity and a size equal to or greater than four pixels (0.36 μm) were counted as a function of nerve area in normal and lesioned IONs. These measurements show that lesioned IONs demonstrate a significantly (* p ≤ 0.0001) higher density of these accumulations per mm2 area than that seen in the normal IONs.
Fig. 10: The average size of sodium channel (NaCh) accumulations increases in lesioned infraorbital nerves (IONs). The mean size (in pixels) of NaCh accumulations with maximum pixel intensity and a size equal to or greater than four pixels (0.36 μm2) were determined in normal and lesioned IONs. This analysis shows that the average size of these accumulations in lesioned IONs is significantly (* p ≤ 0.0001) larger than that seen in the normal IONs.
4. Discussion
This study demonstrates that a combined inflammatory and partial axotomy lesion of the ION results in behavior reflecting an increased sensitivity to mechanical stimuli and a dramatic change in the distribution and density of NaChs in single fibers located just proximal to the site of injury. These changes included an increased overall number and size of NaCh accumulations, an increased NaCh immunofluorescence within typical nodes, and an increased occurrence of split nodes, heminodes, and NaCh accumulations at non-caspr sites. These findings may have important implications regarding the contribution of altered NaCh expression within intact myelinated and demyelinating fibers to the development of increased neuronal excitability seen after nerve lesions.
4.1. The model
Others have shown that partial constriction of the ION at the medial aspect of the orbit with chromic suture results in mechanical allodynia within the injured nerve territory (Vos et al., 1994, 1998; Benoliel et al., 2001) and an increased sensitivity to mechanical stimuli was seen here when the chromic suture was applied to the nerve distal to the foramen in combination with a partial axotomy. The lesion takes features from sciatic nerve models of neuropathic pain, including the use of chromic suture from Bennett and Xie (1988), and partial axotomy from Seltzer et al. (1990).
Chromic suture induces an inflammatory lesion (Clatworthy et al., 1995), while another inflammatory source is provided by axotomized fibers undergoing Wallerian degeneration and the release of myelin break-down products that can activate mast cells (Johnson et al., 1988) in the region of the nearby spared and intact fibers. The combination of a chromic suture-induced inflammatory lesion, with direct nerve injury due to partial axotomy, mimics clinical conditions that are commonly seen in the orofacial region and elsewhere. A strength of the model is the spared nerve lesion since neuropathic pain most likely has contributions from both injured and nearby intact nerves (Gold, 2000). One limitation is that the specific reorganization of NaChs within select fibers subjected either to axotomy or those left intact and affected primarily by an inflammatory lesion is not known, but overall the lesion reflects features of those encountered clinically.
4.2. Relationship of NaCh alterations seen in this study to other studies and their origins
The heminodes and split nodes that were commonly seen in lesioned IONs are also seen during node formation (Vabnick et al., 1996), in myelin mutants (Ulzheimer et al., 2004) and especially after the administration of lysolecithin (Dugandzija-Novakovic et al., 1995; Novakovic et al., 1996), where heminodes resulted from segmental demyelination, and split nodes from paranodal demyelination (Arroyo et al., 2004). While some heminodes were seen in normal IONs and may result from either a loss of myelin as A-delta fibers give rise to “naked” nerve endings or even a remodeling of myelin, they were much more common in lesioned IONs. Although some heminodes seen in lesioned IONs may reflect the identification of just one component of split nodes, the overall greater frequency of both split and especially heminodes reflects a significant demyelinating effect of the lesion that may involve a Schwann cell injury.
The “naked” NaCh accumulations were also more common in lesioned IONs. Some of these “naked” clusters may represent accumulations within chronically unmyelinated or demyelinated axons as seen in other studies (England et al., 1990, 1991; Deerinck et al., 1997). One possible origin of “naked” accumulations is from nodal clusters that lose their caspr association after demyelination. This possibility is supported by the finding that the number of maximum NaCh pixels in the “naked” accumulations approached those seen in typical nodes of lesioned IONs and by the loss of caspr in demyelinating axons from patients with multiple sclerosis (Wolswijk and Balesar, 2003). Another possible origin of “naked” clusters is from internodal regions of axons in areas that were never associated with caspr. It is noteworthy that some “naked” NaCh clusters appeared to be maintained even through active demyelination. This finding suggests that NaCh clustering may occur in the absence of Schwann cell contact, even though contact has been identified as a critical event associated with NaCh clustering at nodes (Dugandzija-Novakovic et al., 1995; Novakovic et al., 1996; Ching et al., 1999). This clustering may involve other mechanisms that have not yet been described (Kaplan et al., 1997, 2001).
A novel finding in lesioned IONs was the presence of typical nodes with a greater NaCh immunofluorescence intensity. This augmentation may occur in response to a subtle Schwann cell injury where there is no retraction or only slight retraction in the paranodal region. In this situation some robust nodes may progress to heminodes or even to split nodes if there is a progressive insult to one or two adjacent Schwann cells.
4.3. Demyelination as a basic response to injury
Although the reorganization of NaChs in dysmyelinated fibers has been used to understand the molecular targeting of NaCh to nodes during development and changes associated with NaCh expression at nodes in pathological conditions, the reorganization observed here has important implications regarding demyelination as a basic response associated with axonal injury. Not only does the inflammatory lesion induced by chromic suture result in significant demyelination (Burchiel, 1980), but demyelination can also result from pressure applied to the nerve by a partial constriction injury (Gupta et al., 2004) and as a response to axotomy (Dyck et al., 1985). Demyelination has long been recognized as a central element associated with the pathogenesis of trigeminal neuralgia. Furthermore, the demyelinating diseases multiple sclerosis and Guillain-Barré are both commonly associated with pain (Moulin, 1998). Importantly, demyelination can induce neuropathic pain (Wallace et al., 2003) and it is possible that the reorganization of NaChs at demyelinated sites seen in this study may be involved in this finding.
4.4. Implications of these findings on neuronal excitability
The apparent functional implication of increased NaCh expression within myelinated fibers is one of enhanced neuronal activity. Indeed, ectopic activity of myelinated axons seen after nerve injury can be inhibited with tetrodotoxin (Liu et al., 2001), and increased excitability correlates with the onset of pain behavior seen after nerve ligation (Han et al., 2000; Liu et al., 2000). Other implications of an increased NaCh expression near receptive fields are that decreased thresholds are needed to generate action potentials and these sites may be capable of initiating activity as impulse generators leading to spontaneous activity. The increased expression of NaCh at nodes may also change the electrical characteristics of the action potential, perhaps changing the frequency encoding of nociceptive stimuli through the peripheral and central processes of primary afferents, or in the primary afferent cell body (Amir and Devor, 2003), resulting in an action potential train that more likely results in activation of a second order neuron. Thus the physiological consequences of increased NaCh expression in myelinated fibers could be the generation of either spontaneous or enhancement of stimulus-induced nerve activity.
4.5. Strengths of single fiber analysis studies
The quantitative analysis used allows an evaluation of NaCh expression at isolated sites in single myelinated fibers. Although single fiber analyses are common in dysmyelination and developmental studies that examine the targeting of NaChs to nodes, most studies that have examined the effect of lesions and inflammatory insults on NaCh expression have been done at the cell body level within peripheral sensory ganglia. It is uncertain how changes seen in cell bodies relate to changes within peripheral fibers at the site of injury. At times there may even be a mismatch between cell body and peripheral fiber expression after nerve injury such as reported for the Nav 1.8 isoform (Novakovic et al., 1998). The limitation of the analysis to only those pixels that show maximum intensity is a key feature of the analysis and allows an unbiased comparison of significant NaCh immunofluorescence in critical areas in different samples, with laser gain levels that still allow observation of a wide variation in NaCh pixel intensities. The selection of maximum pixels certainly limits the analysis to a subpopulation of NaChs but at the same time concentrates on that population that is most likely to influence excitability such as those at nodes. The finding that most nodes contained pixels with maximum intensity reflects that functional subpopulations of NaChs were analyzed with this approach. Although this study has demonstrated a change in NaCh expression after a nerve lesion in caspr-associated accumulations and large “naked” accumulations that most likely are also associated with myelinated fibers, the methods did not evaluate NaCh expression in unmyelinated fibers.
5. Summary
This study introduces techniques that allow a systematic evaluation of NaCh localization within accumulations at caspr-identified and non-caspr sites. Results of this analysis demonstrate a robust change in the distribution and density of NaChs in myelinated and demyelinating fibers located just proximal to a combined inflammatory and partial axotomy lesion that results in behavior reflecting an increased sensitivity to mechanical stimuli. Future studies will evaluate the localization of the various NaCh isoforms after different types of nerve injuries in an attempt to further delineate the role of altered NaCh localization in pain states.
Acknowledgements
This work was supported by National Institutes of Health Grant DE 13942 (MAH). We thank Dr. E. Peles for his generosity in providing the anti-caspr antibody.
References
Amir R, Devor M. Electrical excitability of the soma of sensory neurons is required for spike invasion of the soma, but not for through-conduction.
Biophys J. 2003;84:2181-2191.
Arroyo EJ, Sirkowski EE, Chitale R, Scherer SS. Acute demyelination disrupts the molecular organization of peripheral nervous system nodes.
J Comp Neurol. 2004;479:424-434.
Bennett GJ, Xie YK. A peripheral mononeuropathy in rat that produces disorders of pain sensation like those seen in man.
Pain. 1988;33:87-107. [see comment].
Benoliel R, Eliav E, Iadarola MJ. Neuropeptide Y in trigeminal ganglion following chronic constriction injury of the rat infraorbital nerve: is there correlation to somatosensory parameters?
Pain. 2001;91:111-121.
Black JA, Liu S, Tanaka M, Cummins TR, Waxman SG. Changes in the expression of tetrodotoxin-sensitive sodium channels within dorsal root ganglia neurons in inflammatory pain.
Pain. 2004;108:237-247.
Burchiel KJ. Ectopic impulse generation in focally demyelinated trigeminal nerve.
Exp Neurol. 1980;69:423-429.
Ching W, Zanazzi G, Levinson SR, Salzer JL. Clustering of neuronal sodium channels requires contact with myelinating Schwann cells.
J Neurocytol. 1999;28:295-301.
Clatworthy AL, Illich PA, Castro GA, Walters ET. Role of peri-axonal inflammation in the development of thermal hyperalgesia and guarding behavior in a rat model of neuropathic pain.
Neurosci Lett. 1995;184:5-8.
Deerinck TJ, Levinson SR, Bennett GV, Ellisman MH. Clustering of voltage-sensitive sodium channels on axons is independent of direct Schwann cell contact in the dystrophic mouse.
J Neurosci. 1997;17:5080-5088.
DeLeo JA, Yezierski RP. The role of neuroinflammation and neuroimmune activation in persistent pain.
Pain. 2001;90:1-6.
Dugandzija-Novakovic S, Koszowski AG, Levinson SR, Shrager P. Clustering of Na+ channels and node of Ranvier formation in remyelinating axons.
J Neurosci. 1995;15:492-503.
Dyck PJ, Lais A, Karnes J, Sparks M. Peripheral axotomy induces neurofilament decrease, atrophy, demyelination and degeneration of root and fasciculus gracilis fibers.
Brain Res. 1985;340:19-36.
England JD, Gamboni F, Levinson SR, Finger TE. Changed distribution of sodium channels along demyelinated axons.
Proc Natl Acad Sci USA. 1990;87:6777-6780.
England JD, Gamboni F, Levinson SR. Increased numbers of sodium channels form along demyelinated axons.
Brain Res. 1991;548:334-337.
Gold MS. Spinal nerve ligation: what to blame for the pain and why.
Pain. 2000;84:117-120.
Goldin AL, Barchi RL, Caldwell JH, Hofmann F, Howe JR, Hunter JC, et al. Nomenclature of voltage-gated sodium channels.
Neuron. 2000;28:365-368.
Gregg JM. A surgical approach to the ophthalmic-maxillary nerve trunks in the rat.
J Dent Res. 1973;52:392.
Gupta R, Rowshan K, Chao T, Mozaffar T, Steward O. Chronic nerve compression induces local demyelination and remyelination in a rat model of carpal tunnel syndrome.
Exp Neurol. 2004;187:500-508.
Han HC, Lee DH, Chung JM. Characteristics of ectopic discharges in a rat neuropathic pain model.
Pain. 2000;84:253-261.
Hille B., 2001. Ionic channels of excitable membrances, Sinauer, Sunderland.
Jensen TS. Anticonvulsants in neuropathic pain: rationale and clinical evidence. Eur J Pain. 2002;6(Suppl. A):61-68.
Johnson D, Seeldrayers PA, Weiner HL. The role of mast cells in demyelination. 1. Myelin proteins are degraded by mast cell proteases and myelin basic protein and P2 can stimulate mast cell degranulation.
Brain Res. 1988;444:195-198.
Kaplan MR, Meyer-Franke A, Lamber S, Bennett V, Duncan ID, Levinson SR, et al. Induction of sodium channel clustering by oligodendrocytes.
Nature. 1997;386:724-728.
Kaplan MR, Cho MH, Ullian EM, Isom LL, Levinson SR, Barres BA. Differential control of clustering of the sodium channels Na(v)1.2 and Na(v)1.6 at developing CNS nodes of Ranvier.
Neuron. 2001;30:105-119.
Lai J, Hunter JC, Porreca F. The role of voltage-gated sodium channels in neuropathic pain.
Curr Opin in Neurobiol. 2003;13:291-297.
Lai J, Porreca F, Hunter JC, Gold MS. Voltage-gated sodium channels and hyperalgesia.
Annu Rev Pharmacol. 2004;44:371-397.
Liu CN, Wall PD, Ben-Dor E, Michaelis M, Amir R, Devor M. Tactile allodynia in the absence of C-fiber activation: altered firing properties of DRG neurons following spinal nerve injury.
Pain. 2000;85:503-521.
Liu X, Zhou J-L, Chung K, Chung JM. Ion channels associated with the ectopic discharges generated after segmental spinal nerve injury in the rat.
Brain Res. 2001;900:119-127.
McQuay HJ. Neuropathic pain: evidence matters. Eur J Pain. 2002;6(Suppl. A):11-18.
Moulin DE. Pain in central and peripheral demyelinating disorders.
Neurol Clin. 1998;16:889-898.
Novakovic SD, Deerinck TJ, Levinson SR, Shrager P, Ellisman MH. Clusters of axonal Na
+ channels adjacent to remyelinating Schwann cells.
J Neurocytol. 1996;25:403-412.
Novakovic SD, Tzoumaka E, McGivern JG, Haraguchi M, Sangameswaran L, Gogas KR, et al. Distribution of the tetrodotoxin-resistant sodium channel PN3 in rat sensory neurons in normal and neuropathic conditions.
J Neurosci. 1998;18:2174-2187.
Poliak S, Gollan L, Martinez R, Custer A, Einheber S, Salzer JL, et al. Caspr2, a new member of the neurexin superfamily, is localized at the juxtaparanodes of myelinated axons and associates with K
+ channels.
Neuron. 1999;24:1037-1047.
Poliak S, Peles E. The local differentiation of myelinated axons at nodes of Ranvier.
Nat Rev Neurosci. 2003;4:968-980.
Salzer S, James L. Polarized domains of myelinated axons.
Neuron. 2003;40:297-318.
Seltzer Z, Dubner R, Shir Y. A novel behavioral model of neuropathic pain disorders produced in rats by partial sciatic nerve injury.
Pain. 1990;43:205-218.
Suzuki R, Dickenson AH. Neuropathic pain: nerves bursting with excitement.
Neuroreport. 2000;11:R17-R21.
Ulzheimer JC, Peles E, Levinson SR, Martini R. Altered expression of ion channel isoforms at the node of Ranvier in P0-deficient myelin mutants.
Mol Cell Neurosci. 2004;25:83-94.
Vabnick I, Novakovic SD, Levinson SR, Schachner M, Shrager P. The clustering of axonal sodium channels during development of the peripheral nervous system.
J Neurosci. 1996;16:4914-4922.
Vos BP, Hans G, Adriaensen H. Behavioral assessment of facial pain in rats: face grooming patterns after painful and non-painful sensory disturbances in the territory of the rat's infraorbital nerve.
Pain. 1998;76:173-178.
Vos BP, Strassman AM, Maciewicz RJ. Behavioral evidence of trigeminal neuropathic pain following chronic constriction injury to the rat's infraorbital nerve.
J Neurosci. 1994;14:2708-2723.
Wallace VC, Cottrell DF, Brophy PJ, Fleetwood-Walker SM. Focal lysolecithin-induced demyelination of peripheral afferents results in neuropathic pain behavior that is attenuated by cannabinoids.
J Neurosci. 2003;23:3221-3233.
Watkins LR, Maier SF. Beyond neurons: evidence that immune and glial cells contribute to pathological pain states.
Physiol Rev. 2002;82:981-1011.
Wolswijk G, Balesar R. Changes in the expression and localization of the paranodal protein Caspr on axons in chronic multiple sclerosis.
Brain. 2003;126:1638-1649.
Wood JN. Recent advances in understanding molecular mechanisms of primary afferent activation. Gut. 2004;53(Suppl. 2):ii9-ii12.
Wood JN, Boorman JP, Okuse K, Baker MD. Voltage-gated sodium channels and pain pathways.
J Neurobiol. 2004;61:55-71.
