Lithium enhances survival and regrowth of spinal motoneurons after ventral root avulsion
© Fu et al.; licensee BioMed Central Ltd. 2014
Received: 13 February 2014
Accepted: 26 June 2014
Published: 2 July 2014
Skip to main content
© Fu et al.; licensee BioMed Central Ltd. 2014
Received: 13 February 2014
Accepted: 26 June 2014
Published: 2 July 2014
During the clinical treatment of the brachial plexus root avulsion (BPRA), reimplantation surgery can not completely repair the motor function of the hand because the axonal growth velocity of the spinal motoneurons (MNs) is too slow to re-innervate the intrinsic hand muscles before muscle atrophy. Here, we investigated whether lithium can enhance the regenerative capacity of the spinal MNs in a rat model of BPRA.
The avulsion and immediate reimplantation of the C7 and C8 ventral roots were performed and followed with daily intraperitoneal administration of a therapeutic concentrationof LiCl. After a 20 week long-term rehabilitation, the motor function recovery of the injured forepaw was studied by a grasping test. The survival and regeneration of MNs were checked by choline acetyltransferase (ChAT) immunofluorescence and by Fluoro-Gold (FG) retrograde labeling through the median and ulnar nerves of the ventral horn MNs. The number and diameter of the nerve fibers in the median nerve were assessed by toluidine blue staining. Our results showed that lithium plus reimplantation therapy resulted in a significantly higher grasping strength of the digits of the injured forepaw. Lithium plus reimplantation allowed 45.1% ± 8.11% of ChAT-positive MNs to survive the injury and increased the number and diameter of nerve fibers in the median nerve. The number of FG-labeled regenerative MNs was significantly elevated in all of the reimplantation animals. Our present data proved that lithium can enhance the regenerative capacity of spinal MNs.
These results suggest that immediate administration of lithium could be used to assist reimplantation surgery in repairing BPRA injuries in clinical treatment.
Ventral root avulsion is not similar to distal axotomy, which does not cause MN death in adult animals [1–3]. In contrast, avulsion of the spinal roots isolates the MNs from peripheral nerves and glial cells, causing several interrelated damage processes in MNs, including morphological alterations, biochemical disturbances, gene expression dysregulation, metabolic changes and cell deathin the affected spinal cordsegments [1, 4–8]. The vast majority of the corresponding spinal MNs die within 2–6 weeks of injury in adult rats [9, 10], resulting in paralysis of the corresponding muscle groups because the brachial plexus is the only nerve supply to the upper limb. Surgical reimplantation of avulsed ventral roots can rescue the MNs in this model. However, the implantation, which inserts the avulsed ventral rootlets into the parenchyma of the spinal cord, may cause additional damage to the spinal cord and requires more challenging surgical skills. In this study, we employed a new microsurgical technique to restore the connection by positioning the avulsed ventral root on the ventrolateral pial surface of the spinal cord instead of inserting the ventral rootlets into the parenchyma of the spinal cord [4, 11, 12]. However, the growth velocity of the regenerative MN axons is too slow to re-innervate the intrinsic forepaw muscles before the denervated muscle atrophies [13, 14]. We also estimated the effect of drugs on the enhancement of the growth velocity of the regenerative MN axons. Lithium, which is extensively used in the treatment of bipolar mood disorder in clinical settings, has been demonstrated to be neuroprotective against a variety of neuronal insults, such as amyotrophic lateral sclerosis [15–17]. Moreover, lithium has been demonstrated to promote axon regeneration . Our previous study also found that lithium can reinforce the axonal regeneration of rubrospinal tract neurons in chondroitinase ABC treated rats after spinal cord hemisection [19, 20]. However, whether lithium can promote the regeneration of the spinal MNs after ventral root avulsion is still unknown. Here, we tested the effects of therapeutic lithium on the regeneration of spinal MNs of BPRA-injured and reimplantation-treated adult rats.
After the behavioral test, rats were euthanized, and then the spinal cord was carefully dissected to ensure that the C7 and C8 ventral roots had been completely avulsed or successfully re-implanted into the spinal cord. Under a surgical microscope, we found 2–3 intact ventral rootlets of spinal nerves attached to the spinal cord at the anterolateral sulcus in the contralateral C7 or C8 spinal segments (Figure 1C-D). For ventral root reimplantation treated rats, a nerve root was found that was closely attached to the ventrolateral surface of the ipsilateral C7 spinal segment and regrew into the middle trunk of the ipsilateral brachial plexus. Additionally, the other reimplanted ventral root was identified as a nerve that originated from the ventral surface of the ipsilateral C8 spinal segment, and regenerated into the lower trunk of the ipsilateral brachial plexus (Figure 1C). Conversely, for the rats without ventral root reimplantation treatments, there was an obvious gap between the C7 and C8 spinal segments and the trunks of the brachial plexus in avulsion rats (Figure 1D). In all animals of the present study, the C7 and C8 dorsal roots were well connected to the dorsal aspect of the C7 and C8 spinal segments, respectively.
We show here that ventral root reimplantation can protect spinal MNs from death induced by avulsion injury and also elevate the number of regenerated MNs, as well as facilitate growth of the regenerated axon into the median nerve and promote recovery of the grasping function. In addition, systematic administration of lithium reinforces the pro-survival effect of reimplantation surgery and induces a robust increase in the number and diameter of myelinated axons in the median nerve. Moreover, we provided evidence that lithium itself can enhance the survival of injured MNs and regenerate axon regrowth in the median nerve. Here, we used the MN marker ChAT to assess the survival of the affected motor neurons. Previous studies, including ours, have reported that avulsion results in a transient absence of ChAT protein expression inside affected MNs at early post-lesion time points [21–23]. However, ChAT immunoreactivity was easily distinguished, although reduced, in the affected ipsilateral MNs in the present study. We attribute this difference to different time points in the investigation between the previous and the present studies. Previous results emphasized the early post-lesion time period (mostly within 2 weeks), but we counted the ChAT-positive MNs at the end of 20 weeks (late) post-lesion. Our result was supported by the previous study which proved that the ChAT immunoreactive intensity inside the affected lumbar MNs recovers to the normal level at 4 weeks after axonal injury . Additionally, a number of previous studies, which show no significant difference in MN counts between ChAT and Nissl staining  or between ChAT and cresyl violet staining , support our results. Therefore, the number of ChAT-positive MNs can be used as a helpful marker for the survival of affected MNsatlate post-lesion times in axonal injuries of the peripheral nerves.
When repairing root-avulsion injuries, the survival of affected MNs has been confirmed to be of ultimate importance for axonal regeneration . Our present data showed 33% to 40% of cervical MNs survived up to 5 months after reimplantation, which supports the previous studies that showed ventral root reimplantation is beneficial for the survival of the affected MNs [4, 11, 28]. The present data also support the idea that the surgically re-attached nerve acts as a conduit to induce the regenerative axons to regrow towards the peripheral target [11, 13, 29]. We observed a reimplanted ventral root firmly attached to the ventrolateral side of the C7 or C8 spinal cord at the proximal end connected with the ipsilateral middle or lower trunk of the brachial plexus at the distal end. In the present study, the intact C7 and C8 dorsal roots may also contribute to the repositioning of the regenerative ventral roots to the ventral root outlet area of the spinal cord and to the precise reconstruction of the ventral roots within the middle and lower trunks of the brachial plexus. According to previous studies, ventral root combined with dorsal root avulsion induces enhanced glial activation, which may occur in response to the central sensitization of nociceptive neurons at the early stage of the injury , and enhances glial scar formation [31, 32] at the peripheral–central nervous system (PNS-CNS) transition zone at the late stage of the injury. Previously, we have also shown the avulsion induced activation of astrocytes and microglia in both white and gray matter of the injured spinal cord [23, 33]. Therefore, we think that intact dorsal roots, compared to injured dorsal roots, may reduce unfavorable glial activation environments and facilitate the survival of the spinal MNs in the present study.
In addition to reimplantation itself, lithium treatment seems to positively influence the survival of affected MNs. Our data showed that the number of ChAT-positive MNs was higher in the lithium therapy group compared to the animals that only received avulsion or reimplantation. Moreover, there was no significant difference between the Av + Li subgroup and the Re + PBS subgroup in both C7 and C8 ipsilateral ventral horns. This result may imply that lithium has a neuroprotective effect by preventing neuronal death. This is in line with previous studies, for example, in primary cultures of rat cerebellar granule cells and cortical neurons, lithium robustly and potently protect against glutamate-induced, N-methyl-D-aspartate (NMDA) receptor-mediated excitotoxicity .
Given that the nerve fiber diameter distribution is an important method that can contribute to insights into the processes that play a role in nerve regeneration , the current findings suggest that reimplantation increases the number of myelinated axons and greatly ameliorates the phenotype of the lesioned nerve compared the avulsed animals. Furthermore, systemic administration of lithium may reinforce the regrowth-promoting effect of reimplantation surgery on the regenerated myelinated fibers in the median nerve. We also noted that no significant difference was found when comparing the number of myelinated axons in the Av + Li and Re + PBS subgroups, indicating that even lithium alone could protect the lesioned axon from Wallerian degeneration. This result is consistent with previous reports indicating that lithium supports regeneration in the CNS, including RGC axons  and rubrospinal neurons in spinal cord . More importantly, a recent study reported that lithium enhances remyelination of the facial and sciatic nerves .
Equally interesting is our finding that a significant return of the motor function, in the form of digit flexions, in the reimplantation animals began at 4 weeks post-lesion, compared to the avulsion series. Additionally, the reimplantation plus lithium treatment group showed the highest grasp strength among all of the treated animals. This result coincided with the investigation on the spinal cord and median nerve, as well as the previous studies [37, 38]. In addition, the behavioral test demonstrated that the grasping strength in avulsion injured animals treated with lithium was significantly greater than the vehicle group, and this upward trend was even maintained for 20 weeks after treatment. The rats in the two avulsion subgroups without repair surgery should not differ because there was no reimplantation to facilitate regenerating axons to grow towards the target muscle. One possible explanation for the significant effects found here could be that they result from a gradual loss of function in the Av + PBS group due to gradual muscle atrophy, which is prevented in the Av + Li group by the Li treatment. These data are consistent with a previous report that lithium induced an increase in skeletal myotube size by inhibiting GSK3β . Second, anatomical dissection of the rats confirmed that the C7, C8 and T1 ventral roots make up the components of the median and ulnar nerves , and stimulation of the median nerve produces digit flexion in the rats [41, 42]. In the present study, only the C7 and C8 ventral roots were avulsed. The ventral root of T1 was intact, which might contribute to the grasping strength that remained in the avulsed animals. The spinal MN efferent and also the somatosensory afferent from the skin and muscles of the forepaw were proven to be very important during the grasping motor tasks . Our data showed that the ventral root avulsion did not affect the morphology of the corresponding dorsal roots, which indicates the intact C7 and C8 dorsal roots contributed to the somatosensory input of the forepaw and assisted in the functional recovery of digit grasping strength of the avulsed animals.
The combined therapeutic strategy of reimplantation and lithium in the present study resulted in an increase in the survival of the affected cervical MNs, an increase in the number and diameter of median nerve fibers, and finally a better recovery of motor function in the injured forepaw when compared to the single reimplantation surgery. These findingswere supported by a number of recent studies [44–47], which have shown the neuroprotective effect of lithium on neuronal trauma and degeneration [15–17]. Although some of the clinical trials show conflicting effects of the lithium ion regarding altering the progression of cognitive and functional deficits, the results share a consensus on the view that lithium may exert long-term neuroprotective effects [44, 48, 49]. The mechanism by which lithium confers neuroprotection has been widely studied previously. Lithium, as an inhibitor of GSK-3β, can reduce GSK-3 activity during neuronal apoptosis [50–54]. Therefore, lithium might act to reduce the oxidative stress in avulsion injured MNs in the present study because avulsion-induced oxidative stress has been demonstrated to be fatal to the affected MNs [55–57], and GSK3 has been recognized as an intermediate in oxidative stress-induced apoptotic signaling [58–60]. Another important mechanism that may contribute to lithium’s effects on the affected MNs in the present study might be attributed to lithium-induced BDNF activity in the central nervous system [61, 62] because many neurotrophic factors, including BDNF, have been shown to promote survival and enhance regeneration in root-avulsion injuries [63, 64]. Recently, we also found that lithium can enhance the expression of BDNF in NPCs and increases the neuronal differentiation of NPCs in the ipsilateral C7 ventral horns of the avulsion injured rats . During reimplantation-induced regeneration of the MNs, production of BDNF in the distal segment of the spinal nerves and the target are also thought to account for neuronal survival and regeneration [24, 65]. Therefore, lithium-induced BDNF production might also contribute to its effects onthe promotion of survival and enhancement of axonal regeneration in affected MNs in the present study. Lithium-induced activation of Bcl-2 has also been shown to suppress glial scarring and support axon regeneration in retinal ganglion cells [18, 66]. As an inhibitor of GSK3, lithium induced inactivation of GSK3 was shown to promote axon regeneration and facilitate functional recovery after spinal cord transection injury . Our previous study also proved that lithium can facilitate chondroitinase ABC (ChABC) to promote the axonal regeneration of rubrospinal tract neurons in spinal cord hemisection injury [19, 20].
However, our data also showed that the number of FG-labeled MNs was not significantly different between the reimplantation and reimplantation plus lithium treated animals. Moreover, at a location on the median nerve 40 mm distal to the reimplantation site, the number and diameter of myelinated axons was significantly increased in the Re + Li group compared to the Re + PBS group. One possible explanation was that the good morphological structure of regenerating myelinated axons did not guarantee a 100% functional recovery. As for the goal of fully recovering paw grasping function, although we observed regenerated axons in the median nerve in the morphological study, we believe that there is a long way to go, and much more research will be required.
In conclusion, systematic administration of lithium at a therapeutic dose for 20 weeks reinforcesthe pro-survival effect of reimplantation surgery on injured MNs induced by ventral root avulsion. In addition, the combined administration of lithium and reimplantation surgery promotes increases in the number and diameter of median nerve axons, as well as improves the motor functional recovery of digit flexion. The results indicate that lithium has a high potential value in surgeries that repair root-avulsion injuries in clinical settings.
Adult male Sprague–Dawley rats (150–180 g, Laboratory Animal Center of Sun Yat-sen University) were used in the study. The animal experiments were approved by the Committee on the Use of Live Animals for Teaching and Research of the Zhongshan School of Medicine in Sun Yat-sen University. All procedures were performed according to the Chinese National Institutes of Health Guide for the Care and Use of Laboratory Animals. A high standard of care was followed to minimize pain and discomfort to the animals.
Daily intraperitoneal injection of 1.0 mL phosphate buffer solution (PBS) containing 85 mg/kg LiCl (Sigma, St. Louis, MO, USA) was performed for 12 of the avulsion rats (Av + Li, n = 12) and 12 of the reimplantation rats (Re + Li, n = 12). An injection of 1.0 mL of PBS was used as the control for lithium, which was injected intraperitoneally once a day in 12 of the avulsion rats (Av + PBS, n = 12) and 12 of the reimplantation rats (Re + PBS, n = 12). The duration for rehabilitation was 20 weeks, and none of the animals died during that time.
The Grasping tests (GT) were designed to quantify the strength of digit flexion and were described previously . The rat digit flexion of the forepaw was performed by the intrinsic flexor muscles innervated by the median nerve . Briefly, the tail of the rat was gently lifted until only the tested paw grasped a grid connected to an ordinary electronic balance. Then, the rat was lifted further by the tail with the paw firmly grasping the grid. At the moment that the paw lost its grip, the value shown by the electric balance was recorded. Five measurements per forepaw were recorded and the highest value in grams (g) was considered the grasping strength for each rat at each time point. The time interval between each measurement was 5 min. When testing the grasping strength of the ipsilateral paw, the contralateral paws were covered by the adhesive tape before testing.
FG retrograde labeling of the spinal MNs was performed following the procedures described in our previous publications [11, 69]. Briefly, rats were re-anesthetized at 3 days before the end of the 20 week postinjury period. Under a surgical microscope (Leica, Germany), the right median and ulnar nerves above the cubital fossa were exposed and identified. The injection site was approximately 20 ± 5 mm distal to the nerve cut site of the spinal cord. A total of 1 μl of the FG solution (2% w/v, Fluorochrome, Inc., Englewood, CO, USA) was slowly injected under the epineurium into the proximal stumps of the median and ulnar nerves using a micropipette. The injection lasted for approximately 10 seconds, and then the injection site was clamped with microforceps for another 10 seconds to ensure maximal labeling. Finally, the muscle, fascia and skin were sutured successively in layers. The regenerating axons of the C7 and C8 spinal MNs grew into the reimplanted ventral roots, and the number of FG-labeled MNs was quantified to assess the regenerative capacity of the spinal MNs [11, 69]. The FG microinjection through the right median and ulnar nerves of normal rats was used as a control (n = 6) for the FG-labeled MNs.
At the end of the 20 week post-lesion period, all of the animals were administered a lethal dose of chloral hydrate and transcardially perfused with normal saline followed by 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer (PB, pH 7.4). The cervical spinal cord and the brachial plexus were carefully dissected under a microscope to avoid damage to the reimplantation area. The C7 or C8 spinal segments, which were defined as the region between the uppermost and the lowermost roots of the C7 or C8 nerve of the contralateral spinal cord, respectively, together with their dorsal roots and the reimplanted ventral roots were removed. The median nerves, at a location approximately 40 mm distal to the ventral root reimplantation site of the spinal cord, were also harvested. After postfixation by immersion in 4% PFA followed by overnight immersion in 30% (v/v) sucrose in PB solution at 4°C, the transverse sections of the C7 and C8 spinal cord (35 μm), and the median nerves (10 μm) were cut on a cryostat and collected in 0.01 M PBS. Every third section of the spinal cord was used for the investigation of FG-labeled MNs (n = 6) and the ChAT immunofluorescence reaction (n = 6) under a fluorescence microscope. The cross sections of the ventral and dorsal roots were prepared for HE staining (n = 12). Remyelination of the regenerated axons of the MNs was investigated by toluidine blue-stained semithin sections of the median nerves of avulsion (n = 24), reimplantation (n = 24) and normal control (n = 12) animals.
The procedures for ChATimmunofluorescencewere similar to those used in our previous study . Briefly, sections were first washed three times with 0.1 M PBS for 10 min and incubated with 0.3% Triton X-100 and 3% bovine serum albumin (BSA) in 0.1 M PBS at room temperature for 30 min. Then, sections were incubated with goat anti-rat ChAT (1:500, Santa Cruz Biotechnology, Dallas, Texas, USA) for 72 h at 4°C. After washing in PBS, the sections were incubated with tetramethylrhodamineisothiocyanate (TRITC)-conjugated rabbit anti-goat IgG (1:200, Sigma, Saint Louis, MO, USA) at room temperature in the dark for 2 h. After washing in PB, sections were mounted on gelatin-coated glass slides and cover slipped in 0.1 M PBS containing 50% glycerin. The ChAT-positive MNs were examined and counted via a fluorescence microscope.
In the slides containing the ChAT immunofluorescence reactions, MNs showing a cytoplasm stained with the ChAT antibody and a visible nucleus were counted on both sides of each C7 and C8 ventral horn under a 20× objective lens. Previous studies have demonstrated that both ipsilateral and contralateral ventral horn MNs are ChAT-positive . Therefore, the number of ChAT-positive MNs on the contralateral side was used as an internal control for each section and expressed as 100%. The number of ChAT-positive MNs in the ipsilateral ventral horn, which indicated the number of surviving MNs, was expressed as a percentage of that of the contralateral ventral horn in the same section. The mean of the number of ChAT-positive MNs of the 6 rats in each subgroup was recorded as the number of the surviving MNs for each subgroup.
Our previous studies have demonstrated that FG labeling through the unilateral median and ulnar nerves was confined to the ipsilateral ventral horn MNs of the C7 and C8 spinal cord segments . Images of the ipsilateral C7 and C8 ventral horns were captured (10 × and 20 × lens) with a Lucida camera attached to a fluorescence microscope (Carl Zeiss, Germany). Counting of the number of the FG-labeled MNs was performed by two people who were blind to rats’ subgroups, and manual counting was performed as described in our previous studies [5, 23, 71]. The total number of FG-labeled MNs in the ipsilateral C7 and C8 ventral horn were calculated for each rat. The mean of the total number of the 6 rats was recorded as the number of regenerating MNs for each subgroup.
The procedure for processing the semithin sections and toluidine blue staining was described in our previous studies [11, 72]. Briefly, the median nerves were postfixed in 2.5% glutaraldehyde and subsequently in 1% osmium tetroxide. After dehydrating in a series of graded alcohols, the median nerves were cleared in propylene oxide, and embedded in pure Epon. One millimeter semithin sections were cut on an ultramicrotome (Leica, Germany) and mounted on gelatin-coated glass slides. The myelin of the nerve fibers was stained with 1% toluidine blue. Photographs of the morphology of the spinal cord sections were taken under a light microscope (Carl Zeiss, Germany).
The images of toluidine blue-stained sections of control, avulsed and implanted median nerves were captured under a light microscope (Carl Zeiss, Germany) and analyzed using Image-Pro Plus 6.0 (Media Cybernetics, MD, USA) software. The method requires the preliminary software procedures of spatial calibration (micron scale) and setting of color segmentation for quantitative color analysis. Sampling bias was avoided by spreading the micrographs systematically over the entire cross-section, according to the procedure proposed by Mayhew and Sharma [73, 74]. A total of 3–6 sampling fields in each specimen were selected for counting the total number of myelinated axons. For each axon, the axon diameter was calculated, which was used to determine the distribution of axon diameters.
The statistical calculation and data handling were performed by using SPSS software (version 16.0; SPSS, Chicago, IL). All data are expressed as the mean ± S.E.M. (standard error of the mean). The data obtained in the GT and measurement of BW were analyzed by one-way repeated measures analysis of variance (RM-ANOVA). The test applied to the values from the different time-point assessments followed by Bonferroni post hoc comparisons. The quantification of spinal MNs, including the ChAT immunoreaction (IR), FG retrograde labeling, and median nerve regrowth evaluation by toluidine blue staining, were performed by one-wayANOVA followed by Bonferroni post-hoc comparisons between different treatments. A p value < 0.05 was considered to be statistically significant.
Brain-derived neurotrophic factor
Brachial plexus root avulsion
Bovine serum albumin
Glycogen synthase kinase-3
Phosphate buffer solution
This work was supported by grants from the Department of Science and Technology of Guangdong Province (2010B031600037) and the National Natural Science Foundation of China (81070995). The authors declare no conflicts of interest.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.