Cytokine Prole and Glial Activation Following Brachial Plexus Roots Avulsion Injury in Mice

Background Inammation and tissue inltration by various immune cells play a signicant role in the pathogenesis of neurons suffering the central nervous systems diseases. Although brachial plexus root avulsion (BPRA) leads to dramatic motoneurons (MNs) death and permanent loss of function, however, the crosstalk between cytokines and glial reaction in the spinal cord during injury is far beyond our knowledge. The current study is sought to investigate the alteration of specic cytokine expression patterns of BPRA injured spinal cord during an acute and subacute period. Methods Here, cytokine assay, transmission electron microscopy, and histological staining were utilized to assess the alteration of cytokine network, ultrastructure morphology, as well as glial activation and loss of MNs within two weeks, post-injury on a mouse BPRA model. Results We found that BPRA signicantly changed the level of CXCL1, IL12 p70, ICAM1, IP10, MCP-5, MIP1-α and CD93. Moreover, the elevated MIP1-α and CD93 were mostly distributed in MNs of the injured, but few in healthy segments. Also, BPRA signicantly induced glial reactions in the ventral horn of injured spinal segments, reected by robustly elevated the Ionized calcium-binding adaptor molecule 1 (IBA1) and Glial brillary acidic protein (GFAP) positive cells. We also observed a severe progressively loss of injured MNs, with a character of both apoptosis and necrosis. Conclusion Overall, these ndings suggested that the inammatory cytokines associated with glial cell proliferation play a vital role in the pathophysiology MNs death caused by nerve roots injury. The present study showed that BPRA substantially caused MNs loss in the ventral horn of the injured spinal cord. The damaged MNs is morphologically characteristic of swelling and depletion of mitochondria endoplasmic reticulum, and eventual rupture of organelles, and disruption of nuclear and plasma membranes. These changes are associated with glial activation in the affected ventral horn, reected by enhanced IBA-1 and GFAP immunoreactivity. Moreover, we also found that some essential cytokines level substantially changed during the acute and subacute phase of MNs death, especially like MIP1-α and CD93 expression increased in the inured MNs. These data strongly suggest that cytokines might have crosstalk with glial activation, which contributes to the spinal MNs loss caused by avulsion injury. Our study the changes of cytokine prole following BPRA (acute and subacute periods) in mice compared with contralateral ventral horns in the same injured spinal segments. During the rst 2 weeks after BPRA, we detected signicant changes in concentrations of 7 cytokines with different inammatory and immunological activities. We found CXCL1 showed a rapidly enhanced concentration on 1 dpi then decreased to the baseline during 2–14 dpi. CXCL1, being a chemotactic substance for neutrophils, is expressed by macrophages, neutrophils, astrocytes and epithelial cells Studies in mice have shown that CXCL1 reduces the severity of multiple sclerosis and can exhibit a neuroprotective function. The concurrence of CXCR2(the receptor for CXCL1) on oligodendrocytes


Background
Motoneurons (MNs) degenerate progressively and nally die after the brachial plexus root avulsion (BPRA), an injury-causing paralysis of the target muscle groups of the upper limbs [1] [2].
Neuroin ammation is one of the essential factors for triggering the neuron loss related to acute or chronic neurodegenerative disorders [3]. It has been well documented that the neuroin ammation also plays a causal role for the MNs degeneration and death associated with spinal cord injury, including avulsion injury [4][5] [6] [7].
It has been reported that the neuroin ammation is characteristic of the in ltration of leukocytes [8] and glial activation during the acute and subacute period after spinal cord injury [9] [10] [11].
The in ammation of BPRA is attributed to that lesion created cellular debris, which activates the astrocytes and microglia, trigging the release of a wide variety of oxidative stress regulators, growth factors, and in ammatory mediators including cytokines [12]. The cytokines are proteins that coordinate the immune response throughout the CNS, recruiting phagocytic cells, like peripheral neutrophils and macrophages into the injured spinal cord [13] [14], which are considered to be the rst wave of in ltrating immune cells [15].
The current study is sought to investigate how the alteration of speci c cytokines expression patterns affects spinal glial activation and MNs death induced by BPRA. We observed that substantially elevated glial reactions and enhanced levels of cytokines, especially MIP-1α, which might contribute to a progressively MNs loss in the ventral horn of injured spinal segments caused by avulsion injury.

Animals
Experiments were conducted on 76 adult male C57BL/6 mice, 8-10 weeks old, weighing 20-25 g, which were purchased from the Laboratory Animal Center of Sun Yat-sen University (Guangzhou, China). The number of the animal use permit is SYXK 2017-0081. The mice were housed under a 12-hour light/dark cycle with food and water were available ad libitum. Surgical and animal care procedures were carried out under the provisions outlined in the National Health and Medical Research Council animal ethics and ARRIVE guidelines. All procedures were performed with the approval of the Animal Care and Utilization Committee of Sun Yat-sen University.

Animal Model Setting
The microsurgery procedure for the brachial plexus roots avulsion injury was as we described previously [32] [33] [34][35].In short, the mice were under anesthesia induced with ketamine/xylazine (80/20 mg/kg, i.p.) and maintained with 1% iso urane. With each mouse in the supine position, the right side of brachial plexus was exposed, and its C5-T1 spinal nerve roots were separated under a surgical microscope (ChengHe Microsurgical Instruments Factory, Ning Bo, China). Both dorsal and ventral rootlets were pulled out using micro hemostatic forceps. All the avulsed distal parts of roots were cut away, and the success of the avulsion model con rmed under the microscope. The muscle and skin were sutured in successive layers. The mice were placed in a heated recovery chamber until they were fully awake and then returned to their cages.

Tissue Preparation
To assess the spinal cord cytokines pro le of BPRA mice, a serial of time points [1, 2, 3, 5, 7 and 14 day post injury (dpi)], mice were anesthetized with 1% sodium pentobarbital (40 mg/kg, i.p., Sigma-Aldrich) and the cervical spinal cord was exposed by laminectomy under the surgical microscope. Subsequently, the C5 to T1 spinal segments were quickly removed and immediately divided into ipsilateral and contralateral halves. Spinal cord tissues were frozen in liquid nitrogen for subsequent protein extract. The frozen samples were homogenized in Whole-Cell Lysis buffer (KGP2100, Key GEN Biotech. Co., Ltd. Nanjing, China) containing protease inhibitor and 1 mM PMSF (P7626; Sigma-Aldrich, St. Louis, Missouri) using an electric homogenizer. The homogenates were lysed in ice for 1 h and centrifuged at 12,000 rpm for 30 min at 4 °C. The supernatants were collected, and the protein concentrations were measured using the BCA (SL260619, Thermo Scienti c; Rockford, Illinois) method and stored at − 80 °C until the time of cytokine measurement.

Immuno uorescence, Neutral Red Staining And Cell Counting
Twenty BPRA mice were sacri ced using a lethal dose of 1% sodium pentobarbital on the 1, 3, 7, and 14 dpi, respectively (n = 5/group at each time point). They were transcardially perfused with normal saline followed by cold 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer (PB, pH 7.4). The cervical spinal cord was exposed by laminectomy under the surgical microscope. The C7-C8 spinal segments, which were de ned as the region between the uppermost and the lowermost rootlets of the contralateral C7-C8 spinal nerves, were post-xed 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-C8 spinal cord (30 µm), were collected in 0.01M PBS. Every third section of the spinal cord was used for staining study.
In addition, the neutral red staining was performed in C7 and C8 segments to assess MNs survival as previously reported [36][37]. Brie y, Sections were stained with 1% neutral red (N4638, Sigma-Aldrich, USA) in 0.1M acetic acid (pH4.8) for 2 h followed by dehydration in ethanol. Images were captured using a digital camera attached to the microscope. Next, the quanti cation of cell counting on the ventral horns of the both sides spinal cord (C7-C8 segments Rexed's lamina IX) was as described previously [38]. We computed the mean number of either IF or neural red positive cells from the total 7 ~ 8 sections obtained from each mouse. Only those neurons with both the nucleolus in the nucleus and Nissl bodies in the cytoplasm stained with neutral red we recounted under a 20 × objective lens. Two independent persons blinded to the sidedness of the groups performed cell counting, pooling of means and data analysis.

Transmission Electron Microscopy
Transmission electron microscopy (TEM) was used to assess motor neuron morphology in the spinal cord after BPRA at 1, 3, 7, 14 dpi (n = 5 mice/group at each time point). The mouse was perfused transcardially with 0.9% normal saline, followed by a mixture of 2% PFA and 2.5% glutaraldehyde (Sigma-Aldrich, G6257, USA) in 0.1 M PBS. Approximately 1 mm 3 of tissue per mice was dissected from the spinal cord and xed in 2% glutaraldehyde for 2 h at 4 °C. The tissues were rinsed in 0.1 M cacodylate buffer and post-xed with 1% osmium tetroxide for 2 h. Then, the tissue was rinsed with distilled water before undergoing dehydration in a graded ethanol series. Subsequently, the tissue was in ltrated overnight at 4 °C using a mixture of half acetone and half resin. The tissue was embedded in resin 24 h later and then cured fully as follows: 37 °C overnight, 45 °C for 12 h, and 60 °C for 24 h. After that, 70-nm sections were cut and stained with 3% uranyl acetate for 20 min and 0.5% lead citrate for 5 min. Ultrastructural changes motor neuron morphology was observed under TEM.

Statistical Analysis
All data were expressed as mean ± SEM (standard error of the mean). The morphological evaluation result or cytokine assay result was subjected to one-way or two-way measures ANOVA followed by a Bonferroni or Tukey post hoc test for multiple group comparisons. A value of p < 0.05 was considered statistically signi cant.

BPRA substantially caused MNs loss in the ventral horn of the injured spinal cord
At 1 dpi, the injured MNs started to present morphological features of both apoptosis and necrosis, re ected by slightly swollen mitochondria and condensed cytoplasm with a still visible nucleus and nucleolus. By 3-7 dpi, the cytoplasm of MNs had become hyperchromatic, the nuclear envelope was indistinct and irregular in shape, and the nucleoli and nuclei were more condensed. The cytoplasm also contained fragmented organelles, numerous, small, granular pro les and organelles are signi cantly reduced in the cytoplasm. Mitochondria were absent, and the cell membrane was disrupted (Fig. 1).
Moreover, the light microscopic observation also revealed an increase in the size of the cytoplasm of To evaluate the glial reactions that occurred in the injured spinal cord after BPRA, the expression of IBA-1 and GFAP, markers of activated microglia and astrocytes, were determined using immunoreactivity 1 to 14 dpi in both ipsilateral and contralateral side of the spinal ventral horn. Our results showed both IBA-1 average uorescence intensity (AFI) and immunoreactivity (IR)-positive cells number of ipsilateral ventral horn gradually increased compared to the contralateral side within 1-14 days followed injury (AFI: F (3,16) = 487.7, p < 0.0001; IR cell number, F (3,16) = 249.9, p < 0.0001, Fig. 3). Similarly, we observed a robustly elevated GFAP expression trend in the injured ventral horn after BPRA. Compared to the contralateral side, BPRA caused an increase in GFAP AFI ( F (3,16) = 85.06, p < 0.0001) and IR cell number (F (3 16) = 52.7, p < 0.0001) in the ipsilateral side, with the peaks at 7 dpi (Fig. 4). These data suggested that BPRA produced microglia and glial activation in the injured spinal cord.

Discussion
The present study showed that BPRA substantially caused MNs loss in the ventral horn of the injured spinal cord. The damaged MNs is morphologically characteristic of swelling and depletion of mitochondria endoplasmic reticulum, and eventual rupture of organelles, and disruption of nuclear and plasma membranes. These changes are associated with glial activation in the affected ventral horn, re ected by enhanced IBA-1 and GFAP immunoreactivity. Moreover, we also found that some essential cytokines level substantially changed during the acute and subacute phase of MNs death, especially like MIP1-α and CD93 expression increased in the inured MNs. These data strongly suggest that cytokines might have crosstalk with glial activation, which contributes to the spinal MNs loss caused by avulsion injury.
Our study the changes of cytokine pro le following BPRA (acute and subacute periods) in mice compared with contralateral ventral horns in the same injured spinal segments. During the rst 2 weeks after BPRA, we detected signi cant changes in concentrations of 7 cytokines with different in ammatory and immunological activities. We found CXCL1 showed a rapidly enhanced concentration on 1 dpi then decreased to the baseline during 2-14 dpi. CXCL1, being a chemotactic substance for neutrophils, is expressed by macrophages, neutrophils, astrocytes and epithelial cells [40]. Studies in mice have shown that CXCL1 reduces the severity of multiple sclerosis and can exhibit a neuroprotective function. The concurrence of CXCR2(the receptor for CXCL1) on oligodendrocytes and induced CXCL1 on hypertrophic astrocytes in MS provides a novel mechanism for recruitment of oligodendrocytes to areas of damage, an essential prerequisite for lesion repair in this devastating condition. [41] [42]. CXCL1 of might be actively involved in the regenerative processes by recruiting oligodendrocytes to the injured spinal cord on the rst 1 day after BPRA.
On 2 dpi, we observed a reduced concentration of IL-12 p70 and maintained a relatively low level as compared to 1 dpi. Interleukin-12 (IL-12) is a heterodimeric cytokine produced mostly by phagocytic cells in response to bacteria, bacterial products, and intracellular parasites, and to some degree by B lymphocytes. The early preference expressed in the immune response depends on the balance between IL-12, which favors Thl responses, and IL-4, which favors Th2 responses. Thus,IL-12 represents a functional bridge between the early nonspeci c innate resistance and the subsequent antigen-speci c adaptive immunity [43].
Pro-in ammatory cytokine IP10 peaked at 1 dpi with a second peak at 14 dpi compared to the contralateral side of the spinal cord, which is secreted by several cell types (e.g., leukocytes, macrophages and endothelial cells) in response to IFN-γ and recruits these cells to sites of infection or in ammation after binding to C-X-C motif receptor (CXCR3) [44] [45]. IP10 is expressed at low levels in the spinal cord under normal conditions, but upon stimulation, the expression of this chemokine is substantially increased at sites where glial cells accumulate [45]. Previous study found the neutralization of CXCR3 signi cantly reduces the expression of markers of activated microglia and astrocytes as well as the levels of proin ammatory cytokines and chemokines, such as CCL2 and CXCL10 [45] [46][47].
Changes in the level of pro-in ammatory cytokine MCP5 in the form of a sharp increase in its level on 1 dpi were observed in injured spinal cord and maintained a relatively high level as compared to the contralateral side till 14 dpi. The cytokine MCP5 attracts eosinophils, monocytes, and lymphocytes but not neutrophils. Previous study found sequence analyses identi ed HIF-1α binding sites in the promoters of MCP-5 genes and HIF-1α is involved in transcriptional induction of MCP5 in astrocytes by hypoxia [48].
We found the pro-in ammatory chemokine MIP-1 α during the rst 2 weeks after BPRA with a maximum increase of 13.94 times on 1 dpi, being a macrophage in ammation protein, activates granulocytes, which leads to acute neuroin ammation [49]. Immuno uorescence staining found MIP-1 α mostly located in the injured motor neuron after BPRA. Its increase in the spinal cord tissues may be associated with disruption of the blood-brain barrier and an increased migration of macrophages and neutrophils to the site of injury. Previous studies have shown the importance of MIP-1α in demyelination in the CNS and highlight its effect by exposing MIP-1α knockout mice (MIP-1α-/-) to cuprizone and comparing pathology to wild-type mice, which found demyelination was signi cantly decreased in MIP-1α-/-mice (near 36% reduction) [50]. We speculate that the high expression of MIP-1α may be related to attract macrophages in ltration and demyelination of motor neurons after brachial plexus avulsion in mice.
On 1 dpi, we observed a rapidly enhanced concentration of CXCL1 in the injured spinal cord. Studies in mice have shown that CXCL1 reduces the severity of multiple sclerosis and can exhibit a neuroprotective function. The concurrence of CXCR2(the receptor for CXCL1) on oligodendrocytes and induced CXCL1 on hypertrophic astrocytes in MS provides a novel mechanism for recruitment of oligodendrocytes to areas of damage, an essential prerequisite for lesion repair in this devastating condition. [41] [42]. CXCL1 of might be actively involved in the regenerative processes in the CNS by recruiting oligodendrocytes to the injured spinal cord on the rst 1 day after BPRA.
CD93 is a type I transmembrane glycoprotein that is expressed on the surface of platelets, and myeloid, endothelial, hematopoietic stem and natural killer cells. CD93 is strongly associated with immune cell phagocytosis, which regulates innate immunity and in ammation [51]. CD93 deletion also induces phagocyte impairment on scavenging apoptotic cells [52]. Our study found CD93 IR was low in the contralateral side of spinal cord, but increased during in ammatory reaction in the ipsilateral side of spinal cord, reaching a peak at early stage of BPRA and showing distinct cytoplasmic and nuclear staining. These ndings suggest that the CD93 cytoplasmic tail is mainly expressed in cell membranes yet may migrate towards the cytoplasm or nucleus with increasing in ammatory reaction.
The levels of the ICAM1 gradually increased with a peak reached on 14 day. As a cell-surface glycoprotein typically expressed by both immune system cells and endothelial cells. ICAM1 is also expressed by astrocytes, though at low levels in resting, quiescent astrocytes, remarkably it is induced 12-fold following injury and subsequent astrocyte activation [53]. ICAM1 may facilitate astrocyte-lymphocyte interactions ultimately aiding recruitment of immune cells into the CNS [54].

Conclusions
In our study, we evaluated the cytokine pro le in spinal cord of mice with a model of BPRA in acute and subacute periods after injury. Our study also demonstrated that the MNs degeneration process, in ammatory reaction and immunological response after BPRA. A further study of a complex cytokine network imbalance after BPRA and determining its possible correlation with the severity of damage and clinical events seems important. This would provide a better understanding of the role cytokines play in the pathophysiology of the BPRA disease. The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval
All animal experiments were approved by the Animal Care and Use Committee of Sun Yat-sen University and the Guangdong Province Animal Care Ethics Committee.

Consent for publication
Not applicable.  Unilateral BPRA caused a dramatic motoneurons (MNs) loss in the ventral horn of the injured spinal cord.
The spinal cord was harvested for cryostat, the 30μm section counterstained with neutral red at 1-14 days post-injury (dpi) to evaluate the MNs survival rate. Panel A is the representative microphotographs that depict the coronal section of spinal segments suffered from unilateral BPRA at 1-14 dpi (a-d). a1-d1, a2-d2 are representative images showing MNs in the contralateral (Con) and ipsilateral (Ipsi) ventral horn of the spinal cord (outlined in the rectangular area in a-d) respectively. Panel B summarizes the change of the MNs survival rate during 1-14 dpi. ****p< 0.0001 vs. 1d, One-way ANOVA followed by Tukey post hoc test, n=5 mice/group. All data were presented in mean ± S.E.M. The white arrow indicates motoneurons.