Neuroprotection and immunomodulation after treatment with NeuroBoost and NeuroHeal following ventral root crush in mice
Lilian de Oliveira Coser, Maria Fernanda Vannucci, Júlia Lombardi, Luciana Politti Cartarozzi, Alexandre Leite Rodrigues de Oliveira
TL;DR
Two drug combinations, NeuroBoost and NeuroHeal, were tested in mice for their ability to protect neurons and reduce immune responses after spinal nerve injury.
Contribution
The study compares the neuroprotective and immunomodulatory effects of NeuroBoost and NeuroHeal in a mouse model of spinal nerve injury.
Findings
NeuroBoost improved motoneuron survival and reduced glial and immune responses after injury.
NeuroHeal also enhanced neuronal survival and reduced astrogliosis, with increased M1 macrophages.
Both treatments supported functional recovery in mice after spinal nerve injury.
Abstract
Pharmacological immunomodulation can prevent neuronal loss and reactive gliosis after injuries to spinal nerve roots. NeuroBoost (NB) and NeuroHeal (NH) were used to compare their effects on neuronal survival, glial and immune response, and functional recovery after root injury in mice. Our data showed that the NB combination presented a significant difference in the survival of motoneurons after injury compared to the vehicle group (0.71 ± 0.07 vs. 0.60 ± 0.01, p = 0.04, ratio ipsi/contralateral), decreased astrogliosis and microglia reactivity (astrogliosis: 7 dpi: VE vs. NB 4.4 ± 1.06 vs. 3.08 ± 0.39, p = 0.007; 28 dpi: VE vs. NB 5.02 ± 1.06 vs. 1.82 ± 0.53, p = 0.0003; microglial reactivity: 28 dpi: VE vs. NB 3.95 ± 0.88 vs. 2.63 ± 0.37, p = 0.02; integrated density of pixels – ratio ipsi/contralateral). A shift to a non-reactive profile was observed and functionally, the NB…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsNerve injury and regeneration · Neurogenesis and neuroplasticity mechanisms · Pain Mechanisms and Treatments
Introduction
Spinal cord injuries (SCI) may happen after brachial and lumbosacral plexuses trauma, as frequently caused by motor vehicle accidents, falls, violence, and sports ^1,2^. Such lesions lead to temporary or permanent loss of motor, sensory, and autonomic functions of spinal cord^3^. They also result in motoneuron degeneration, astroglial and microglial persistent activation, and loss of sensory and motor function (Mazzer et al., 2008, Kobayashi et al., 2004, Anderson, 2004, Thuret et al., 2006).
Motor root injury results in paralysis of the ipsilateral limb, and spontaneous recovery is unlikely, although different surgical approaches can minimize the long-term sequelae. Based on the fact that spinal nerve roots may be crushed due to spondylolisthesis, disc herniation, spinal stenosis, and tumors, we have proposed an animal model to recapitulate such proximal axotomy (Mazzer et al., 2008, Kobayashi et al., 2004). In this sense, we demonstrated that, although less aggressive than avulsion, root crush results in death of motoneurons, combined with microglial and astrocyte reactions. Further, it has also been demonstrated loss of inputs to motoneurons (Spejo et al., 2013), so that there is need of developing strategies to minimize loss of neurons and further degeneration of spinal circuitry.
During synaptic loss following spinal cord injury, microglia and astrocytes play a crucial role. They are rapidly activated after injury, extending their processes between retracted synaptic terminals and the postsynaptic membrane of injured motor neurons (Aldskogius, 2011, Cullheim and Thams, 2007, Oliveira and Langone, 2000, Rotterman and Alvarez, 2020, Salvany et al., 2021). Reactive astrocytes can be categorized into two phenotypes with distinct functional profiles. A1 astrocytes are associated with neuroinflammation and are induced through activation of the NF-κB signaling pathway, leading to the secretion of pro-inflammatory cytokines and complement system components such as C1q. These molecules, when present in the central nervous system (CNS) microenvironment, contribute to neuronal damage (Liddelow et al., 2017, Bush et al., 1999, Zamanian et al., 2012). In contrast, A2 astrocytes are activated predominantly via the STAT3 signaling pathway and are characterized by the release of neurotrophic factors that support neuronal survival and axonal growth, thereby exhibiting a neuroprotective profile (Anderson et al., 2016, Gao et al., 2005, Hayakawa et al., 2014, Lian et al., 2015, Zador et al., 2009).
Similarly, macrophages/microglia exhibit distinct functional profiles—M1 (pro-inflammatory) and M2 (anti-inflammatory)—with additional M2a, M2b, and M2c subtypes, depending on the cytokine release in the environment. M1 macrophages are driven by IL-6, IFN-γ, IL-12, IL-23, IL-1β, and TNF-α, while M2 produce IL-10 and TGF-β and suppress NF-κB-mediated signaling, aiding in inflammation resolution (Anwar et al., 2016, Durafourt et al., 2012, Gensel and Zhang, 2015, Kroner et al., 2014, Sica et al., 2008, Zhou et al., 2014). In synergy with glial cells, the early post-injury surge of TNF-α, IL-1, and IL-6 leads to the chemotaxis of neutrophils, macrophages, and T lymphocytes, which further amplify immune cell recruitment at the lesion site (Pineau and Lacroix, 2007, Sánchez-Ventura et al., 2019, Zhou et al., 2018).
Neuroprotective and immunomodulatory drugs need to be investigated in the context of root crush injury aiming to decrease the inflammatory process and neuron death. NeuroBoost is composed of a combination of drugs, 4-hydroxy-TEMPO, and dimethyl fumarate (DMF). Some studies have described the antioxidant and anti-inflammatory functions of these drugs. Both molecules cross the blood-brain barrier (Behringer et al., 2002, Gilgun-Sherki et al., 2002, Mitchell et al., 1990), and may be administered orally with low side effects.
Spejo and colleagues demonstrated, by using the root crush model, that 4-hydroxy-TEMPO reduces inflammation, controls microglial and astroglial reactions leading to spinal cord circuitry preservation (Spejo et al., 2019). Other studies from our lab obtained significant protection of motoneurons, decreased glial response, and improved life span in SODG93A mice, a model of amyotrophic lateral sclerosis, and after neonatal sciatic nerve transection (Chiarotto et al., 2014).
DMF plays an immunomodulatory role in autoimmune diseases, such as multiple sclerosis, in which it has promoted a reduction in microglial reactivity and macrophages migration, further preservation of myelin and axonal integrity (Bomprezzi, 2015). The immunological role of this drug consists of the switch to Th2 response, associated with IL-4 and IL-5 production (Albrecht et al., 2012). Further, the immunomodulatory role of this medication has also been described by leading to an improvement of neuronal survival, and protection from oxidative damage. Administration of DMF increases the level of neurotrophic factors, like BDNF, GDNF, and NT3 (Cordaro et al., 2017).
NeuroHeal (acamprosate and ribavirin), has been recently described by Romeo-Guitart and colleges (2018), and they have demonstrated that the combination promoted neuroprotection, nerve regeneration, and functional recovery after injury. Importantly, both acamprosate and ribavirin cross the blood-brain barrier, reducing gliosis, and leading to upregulation of growth associate protein 43 (GAP-43) (Romeo-Guitart et al., 2018a). After root avulsion in rats, NeuroHeal treatment led to long-lasting neuroprotection that remained after six months of treatment (Romeo-Guitart et al., 2017). Acamprosate is used for alcoholism treatment, interacting with GABA, NMDA and mGlut5 receptors (Kalk and Lingford-Hughes, 2014, De Witte et al., 2005, Harris et al., 2002, Rammes et al., 2001). Ribavirin, on the other hand, is used for hepatitis C treatment, although it has CNS effects such as reducing infiltration of CD4 and CD8 T cells, and decrease of gliosis (Solbrig et al., 2002, Lee et al., 2008).
In this study, we evaluated the effects of the combinatory treatments NeuroBoost (dimethyl fumarate and Tempol) and NeuroHeal (acamprosate and ribavirin), both of which have shown promising neuroprotective and immunomodulatory properties in the context of spinal cord injury. To further characterize the inflammatory response, we employed flow cytometry to assess the polarization of glial cells, such as astrocytes and microglia, as well as immune cells, including lymphocytes and macrophages, following ventral root crush injury and treatment with the respective combinations. This approach allowed a comprehensive evaluation of the inflammatory process at the lesion site across different experimental time points. By employing a multifaceted strategy, we were able to broadly assess the neuroprotective potential of these treatments, alongside synaptic preservation, functional motor recovery, and glial reactivity and polarization following injury and therapeutic intervention.
Results
Neuroboost and Neuroheal presents neuroprotective role after ventral root crush
NeuroHeal combination led to increased neuronal survival at 7 days post-injury, with a survival rate of 77 % compared to 62 % in the vehicle-treated group (0.77 ± 0.02 vs. 0.62 ± 0.04; p = 0.04). After 14 days, no statistically significant difference was observed, with a survival rate of 66 % in the NeuroHeal group compared to 60 % in the vehicle group. Similarly, at 28 dpi, the NeuroHeal-treated group exhibited a survival rate of 53 %, which was not significantly different from the 54 % observed in the vehicle group (Fig. 1).Fig. 1Neuronal Survival. Survival of spinal motoneurons after ventral root crush injury in C57BL/6 J mice. Contralateral (uninjured) side at 7 days (A), 14 days (B), and 28 days post-injury (C). Vehicle-treated animals at 7 days (D), 14 days (E), and 28 days post-injury (F). NeuroBoost-treated animals at 7 days (G), 14 days (H), and 28 days after treatment initiation (I). NeuroHeal-treated animals at 7 days (J), 14 days (K), and 28 days after treatment initiation (L). Representative sections of the spinal cord showing Rexed lamina IX (squares) on the ipsilateral (IL) and contralateral (CL) sides used for motoneuron quantification (M). Ipsilateral/Contralateral motoneuron survival ratio in Rexed lamina IX of the spinal cord (N). One-way ANOVA – t test, p < 0.05 (*).Fig. 1
In contrast, treatment with NeuroBoost did not yield a significant difference at 7 days post-treatment, with a survival rate of 64 % versus 62 % in the vehicle group. However, after 14 days, significant neuronal survival was observed in the NeuroBoost-treated group (71 %) compared to the vehicle group (60 %) (0.71 ± 0.03 vs. 0.60 ± 0.005; p = 0.04). At 28 days post-treatment, no significant difference was observed, with the NeuroBoost group showing a survival rate of 61 % compared to 54 % in the vehicle group (Fig. 1).
Reduction in microglial and astroglial reaction after treatment with both combinations
Quantitative analysis of the ipsilateral/contralateral ratio revealed a significant reduction in astroglial reactivity at 7- and 28-days post-injury in animals treated with the NeuroBoost combination (7 dpi: 4.4 ± 0.3 vs. 3.08 ± 0.15, p = 0.007; 28 dpi: 5.02 ± 0.47 vs. 1.82 ± 0.24, p = 0.0003). No statistically significant differences were observed at the 14-day time point (Fig. 2).Fig. 2Astrogliosis. Spinal cord sections following ventral root crush injury at L4, L5, and L6 segments. Contralateral (non-injured) side at all timepoints (A–C) shows low expression of GFAP. In contrast, a strong GFAP signal was detected in the vehicle-treated group at 7 days (D), 14 days (G), and 28 days (J) post-injury. Treatment with NeuroBoost resulted in reduced astroglial reactivity at 7 days (E), 14 days (H), and 28 days (K) post-injury. A similar reduction in GFAP expression was observed in the NeuroHeal-treated group at 7 days (F), 14 days (I), and 28 days (L) post-injury. Triple staining for GFAP (green), NeuN (red), and DAPI was used to visualize Rexed lamina IX (M). Quantification of astroglial reactivity is represented by the integrated pixel density (ipsilateral/contralateral ratio) at each timepoint and condition (N). One-way ANOVA – t test, p < 0.05 (), p < 0.005 (), p < 0.0005 ().Fig. 2
A similar pattern was observed following treatment with the NeuroHeal combination, which also resulted in a significant reduction in astroglial reactivity at 7 and 28 days after treatment initiation (7 dpi: 4.4 ± 0.37 vs. 3.04 ± 0.37, p = 0.031; 28 dpi: 5.02 ± 0.47 vs. 2.27 ± 0.20, p = 0.0007). No significant differences were detected after 14 days post-injury between NeuroHeal-treated animals and the vehicle group (Fig. 2).
Regarding microglial reaction, we observed that after 7 days of treatment, neither of the combinations showed statistically significant differences in microglial reactivity compared to the vehicle group (VE vs. NB: 3.5 ± 0.5 vs. 3.71 ± 0.30, p = 0.81; VE vs. NH: 3.5 ± 0.5 vs. 4.95 ± 0.47, p = 0.09). However, it is worth noting that the NeuroHeal-treated group exhibited significantly higher microglial reactivity than the NeuroBoost-treated group (3.71 ± 0.30 vs. 4.95 ± 0.47, p = 0.04) (Fig. 3).Fig. 3Microglial response – anti-Iba-1 staining. Spinal cord sections following ventral root crush injury at the L4, L5, and L6 segments. Contralateral (non-injured) side at all timepoints (A–C) presents low expression of Iba-1. In contrast, a strong Iba-1 signal was observed in the vehicle-treated group at 7 days (D), 14 days (G), and 28 days (J) post-injury. Microglial reactivity was reduced following treatment with NeuroBoost at 7 days (E), 14 days (H), and 28 days (K) post-injury. A similar pattern was observed in the NeuroHeal-treated group at 7 days (F), 14 days (I), and 28 days (L) post-injury. Triple staining for Iba-1 (red), NeuN (green), and DAPI was used to visualize Rexed lamina IX (M). Quantification of microglial activation is represented by the integrated pixel density (ipsilateral/contralateral ratio) across groups and timepoints (N). One-way ANOVA – t test, p < 0.05 (*).Fig. 3
Conversely, at 14 days post-treatment, we observed an opposite trend, NeuroBoost group showed significantly higher microglial reactivity compared to the NeuroHeal group (5.91 ± 0.85 vs. 3.43 ± 0.64, p = 0.04), although no significant differences were found when compared to the vehicle group.
Interestingly, after 28 days of treatment, the NeuroBoost group exhibited significantly reduced microglial reactivity compared to the vehicle group (3.95 ± 0.39 vs. 2.63 ± 0.28, p = 0.02). This effect was not observed in the NeuroHeal-treated group compared to the vehicle (Fig. 3).
In addition to analyzing microglial reactivity, we also evaluated the structural profile of microglia following ventral root crush injury and subsequent treatment. Our results show that, after 7 days of treatment with the NeuroBoost combination, there was an increased presence of type II microglia (non-reactive profile) in the ipsilateral side of the injury (4.90 % ± 1.43 vs. 23.98 % ± 7.11, p = 0.04) and a significant reduction in type V microglia (reactive profile) compared to the vehicle group (53.87 % ± 2.18 vs. 26.39 % ± 5.98, p = 0.0019) (Fig. 4). A similar increase in type II microglia was observed in the NeuroHeal-treated animals, but in this case, on the contralateral side (56.06 % ± 4.96 vs. 72.31 % ± 10.11, p = 0.04) (Fig. 4). However, on the ipsilateral side, we also observed a decrease in type V microglia in the NeuroHeal group compared to the vehicle (53.86 % ± 2.18 vs. 26.39 % ± 5.98, p = 0.0019) (Table 1; Fig. 4).Fig. 4Microglial structural analysis. Representative photomicrographs of microglial morphological classification based on Iba-1 immunostaining. Type I and II microglia were classified as non-reactive (surveillant microglia), while type III, IV, and V microglia were classified as reactive (activated microglia) (A). Quantitative analysis of the percentage of microglia classified according to their morphological profile at 7, 14, and 28 days post-surgery in the different experimental groups: vehicle group (VE), NeuroBoost group (NB), and NeuroHeal group (NH) (B). Two-way ANOVA - Tukey's multiple comparisons test, p < 0.05 (), p < 0.001 (), p < 0.0001 ().Fig. 4. Table 1Statistical analysis of microglial phenotyping following treatment with vehicle (EV), NeuroBoost (ENB), and NeuroHeal (ENH) at 7, 14, and 28 days post-injury (dpi). Two-way ANOVA, p < 0.05 (), p < 0.001 (), p < 0.0001 ().Table 1Days post-injurySideAverage ± SEMp value**7ILType IIVE x NB4.90 % ± 1.43 vs. 23.98 % ± 7.110.047*CLType IIVE x NH56.06 % ± 4.94 vs. 72.31 % ± 10.110.047ILType VVE x NB53.87 % ± 2.18 vs. 26.39 % ± 5.980.00197ILType VVE x NH53.87 % ± 2.18 vs. 22.58 % ± 3,940.000414ILType IINB x NH40.87 % ± 9.27 vs. 19.80 % ± 3.280.0414*ILType VVE x NB45.70 % ± 8.69 vs. 19.72 % ± 8.950.0114ILType VVE x NH45.70 % ± 8.69 vs. 20.24 % ± 3.470.01**28ILType IVVE x NB34.75 % ± 6.52 vs. 16.80 % ± 3.700.02**28CLType IVE x NB28.37 % ± 2.81 vs. 43.20 % ± 9.720.04IL: Ipsilateral; CL: Contralateral.
In animals treated for 14 days, NeuroBoost led to a significant reduction in type V microglia compared to the vehicle group (45.70 % ± 8.69 vs. 19.72 % ± 8.95, p = 0.01). A similar effect was observed in the NeuroHeal group (45.70 % ± 8.69 vs. 20.24 % ± 3.47, p = 0.01). Additionally, NeuroBoost treatment resulted in a higher percentage of type II (non-reactive) microglia compared to NeuroHeal (40.87 % ± 9.27 vs. 19.80 % ± 3.28, p = 0.04). After 28 days of treatment, NeuroBoost promoted a significant decrease in the percentage of type IV microglia (reactive profile) in the ipsilateral side compared to the vehicle (34.75 % ± 6.52 vs. 16.80 % ± 3.70, p = 0.02), and an increase in type I microglia (non-reactive profile) in the contralateral side (28.37 % ± 2.81 vs. 43.20 % ± 9.72, p = 0.04) (Table 1; Fig. 4). No significant differences were observed in the NeuroHeal-treated group at this timepoint.
The reduction in reactive microglial profiles observed in the NeuroBoost group is consistent with the results obtained through immunofluorescence analysis using anti-Iba-1, which also showed reduced microglial reactivity after 28 days of treatment with the combination**.**
Synaptic preservation after treatment with NeuroHeal
We observed that treatment with the NeuroHeal combination resulted in a significantly higher ipsilateral/contralateral ratio compared to the vehicle group at 7 days post-injury (0.56 ± 0.07 vs. 0.86 ± 0.05, p = 0.01). A similar finding was observed at 28 days post-injury (0.47 ± 0.06 vs. 0.73 ± 0.05, p = 0.04). However, no statistically significant differences were observed between the NeuroHeal and vehicle groups at the 14-day time point (Fig. 5**).**Fig. 5. Synaptic Coverage. Sections of the spinal cord following ventral root crush in the L4, L5, and L6 segments. We observed a higher expression of synaptophysin at the contralateral (uninjured) side (A, B, and C). After injury, a decrease in synaptophysin expression is observed in the vehicle group at different time points: 7 days (D), 14 days (G), and 28 days (J) post-injury. However, with the treatments, there is an increase in synaptophysin expression in the group treated with NeuroBoost at 7 days (E), 14 days (H), and 28 days (K) post-injury. The same observation was made for the NeuroHeal group at 7 days (F), 14 days (I), and 28 days (L) post-injury. Photomicrograph of triple labeling with Synaptophysin (green) + NeuN (red) + DAPI (M). Integrated pixel density (ipsi/contra ratio) of image quantification across different groups and time points (N). One-way ANOVA – t test, p < 0.05 (*).Fig. 5
In contrast, the NeuroBoost-treated group did not show statistically significant differences at any time points when compared to either the vehicle or NeuroHeal groups, displaying values similar to those of the vehicle-treated animals (Fig. 5).
After 7 days of treatment, neither of the combinations resulted in significant differences compared to the vehicle group. However, after 14 days of treatment, a decrease in VGLUT-1 expression was observed in the group treated with the NeuroBoost combination compared to the vehicle group (0.60 ± 0.02 vs. 0.39 ± 0.03, p = 0.01). This finding was also observed when comparing the drug combinations, with the NeuroBoost group showing a lower ipsilateral/contralateral ratio compared to the NeuroHeal group at the same time point (0.39 ± 0.03 vs. 0.54 ± 0.05, p = 0.04).
After 28 days of treatment, both combinations resulted in significantly higher VGLUT-1 expression compared to the vehicle group (VE: 0.39 ± 0.02 vs. NB: 0.63 ± 0.03, p = 0.0009; VE: 0.39 ± 0.02 vs. NH: 0.71 ± 0.01, p < 0.0001), with no significant differences between the combinations (Fig. 6).Fig. 6. Glutamatergic (excitatory) synaptic terminals. Sections of the spinal cord following ventral root crush in the L4, L5, and L6 segments. On the contralateral (uninjured) side, we observed a high expression for VGLUT-1 (A, B, and C). After injury, a decrease in immunoreactivity is observed in the vehicle group at different time points: 7 days (D), 14 days (G), and 28 days (J) post-injury. However, with the treatments, preservation of VGLUT-1 labeling is observed in the group treated with NeuroBoost at 7 days (E), 14 days (H), and 28 days (K) post-injury. The same was observed for the NeuroHeal group at 7 days (F), 14 days (I), and 28 days (L) post-injury. Photomicrograph of triple labeling with VGLUT-1 (green) + NeuN (red) + DAPI (M). Integrated pixel density (ipsi/contra ratio) of image quantification across different groups and time points (N). One-way ANOVA – t test, p < 0.05 (*), p < 0.01 (), p < 0.001 (), p < 0.0001 (***).Fig. 6
Anti-inflammatory profile in astrocytes after 7 days of treatment with NeuroHeal
NeuroHeal combination resulted in an increase in the A1 profile, 33 % vs. 72 %, and A2 profile, approximately 33 % vs. 72 %, after 7 days (A1: 33.27 ± 6.59 vs. 72.88 ± 4.42, p < 0.0001; A2: 39.85 ± 9.04 vs. 82.46 ± 3.05, p < 0.0001) (Figure 13 – D). The higher percentage of the A2 profile was observed specifically in the analysis of IL-4-producing cells, with no significant differences detected in the analysis of IL-10-producing A2 cells (Fig. 7).Fig. 7. Characterization of A1 and A2 astrocytic profiles spinal cord injury. Dot plot graphs illustrate the analysis used to phenotype A1 astrocytes (GFAP⁺TNF-α⁺) and A2 astrocytes (GFAP⁺IL-10⁺IL-4⁺) following treatment with Vehicle, NeuroBoost and NeuroHeal for 7 (A), 14 (B) and 28 days after ventral root injury (C). The percentage of A1 and A2 astrocytes in the respective timepoints in different experimental groups is shown (D – H). A2 astrocytes with double expression of IL-10 +IL-4 + after 7 days (E), 14 days (F) and 28 days post-injury (G). Two-way ANOVA - Tukey's multiple comparisons test, p < 0.05 (*), p < 0.001 (), p < 0.0001 (*).Fig. 7
In addition, we assessed the co-expression of anti-inflammatory cytokines at different time points and treatment conditions through the co-expression of GFAP+IL-10 +IL-4 + . Treatment with the NeuroHeal combination resulted in a higher expression of anti-inflammatory cytokines compared to the vehicle group after 7 days of treatment, with 22 % in the treated group versus 6 % in the vehicle group (6.92 ± 1.75 vs. 22.83 ± 4.23, p = 0.02). Animals treated with NeuroBoost did not show statistically significant differences between groups at the evaluated time points (Fig. 7). Also, we found a decrease in A1 pro-inflammatory profile after 14 days with NeuroHeal treatment (76,66 ± 3,70 vs. 54,41 ± 9,44, p = 0,02) (Fig. 7).
Microglial reaction shows decrease in both M1 and M2 profiles after Neuroheal and Neuroboost treatment
Our data demonstrate that treatment with NeuroBoost resulted in a reduction of the M1 profile after 7 days of treatment compared to the vehicle group, with approximately 89 % in the vehicle group versus 64 % in the NeuroBoost-treated group (89.45 ± 0.54 vs. 64.85 ± 6.69, p = 0.03). No significant difference was observed in the M2 profile (Fig. 8**).Fig. 8. Characterization of M1 and M2 microglia after spinal cord injury. Representative dot plots illustrate the flow cytometry analysis used for the phenotyping of microglia (CD11b⁺CD45^low^), M1 profile (CD68⁺TNF-α⁺), and M2 profile (CD206⁺IL-10⁺) following treatment with vehicle, NeuroBoost, and NeuroHeal after 7 days (A), 14 days (B) and 28 days period after ventral root injury (C). The percentage of M1 and M2 macrophages in the different experimental groups in respective timepoints (D - F). Two-way ANOVA - Tukey's multiple comparisons test, p < 0.05 (), p < 0.0001 ().Fig. 8
A similar finding was observed in the group treated with the NeuroHeal combination, which also showed a reduction in the M1 profile compared to the vehicle group after 7 days of treatment, with 68 % in the treated group (89.45 ± 0.54 vs. 68.91 ± 5.92, p = 0.01). Conversely, the treatment also led to a reduction in the M2 profile compared to the vehicle group, with 27 % in the vehicle group versus 13 % in the treated group (27.67 ± 13.02 vs. 1.10 ± 0.42, p = 0.02). No significant differences between the combinations were observed in either the M1 or M2 profiles at the evaluated time point (Fig. 8).
After 28 dpi, no significant differences were observed in the M1 profile for either combination. However, when evaluating the M2 profile, a decrease was observed in animals treated with the NeuroBoost combination compared to the vehicle group, from 26 % in the vehicle group to 3 % in the NeuroBoost-treated group (26.31 ± 6.96 vs. 3.02 ± 0.82, p = 0.0009). A similar finding was observed in the NeuroHeal group, with approximately 2 % compared to the vehicle group (26.31 ± 6.96 vs. 2.42 ± 1.44, p = 0.0007). No significant differences were found between the combinations in the M2 profile (Fig. 8).
Next, we assessed the inflammatory profile of macrophages in the spinal cord following ventral root crush injury. After 7 days, both treatments did not result in a significant difference in the M1 macrophage profile compared to the vehicle group, although values were highly heterogeneous (Fig. 9**).**Fig. 9. Characterization of M1 and M2 macrophages after spinal cord injury. Representative dot plots illustrate the flow cytometry analysis used for the phenotyping of microglia (CD11b⁺CD45^high^), M1 profile (CD68⁺TNF-α⁺), and M2 profile (CD206⁺IL-10⁺) following treatment with vehicle, NeuroBoost, and NeuroHeal after 7 days (A), 14 days (B) and 28 days period after ventral root injury (C). The percentage of M1 and M2 macrophages in the different experimental groups in respective timepoints (D - F). Two-way ANOVA - Tukey's multiple comparisons test, p < 0.05 (*).Fig. 9
When evaluating the M2 macrophage profile, treatment with NeuroBoost led to a reduction in the percentage of M2 cells compared to the vehicle group, with a decrease from 73 % to 43 % (73.81 ± 10.99 vs. 43.93 ± 3.97, p = 0.04). No significant differences were observed between the NeuroHeal group and the vehicle group (Fig. 9).
Fourteen days post-injury, treatment with the NeuroHeal combination resulted in an increase in M1 macrophage profile compared to the vehicle group, corroborating previous findings (45.26 ± 6.32 vs. 71.23 ± 5.44, p = 0.02). No statistical differences were observed in the NeuroBoost group compared to the vehicle group for the M1 profile (Fig. 9). M2 macrophage profile remained stable after treatment, with no significant differences among the groups at the 14-day time point (Fig. 9).
After 28 days of treatment, no significant differences were observed in either M1 or M2 macrophage profiles in the NeuroBoost and NeuroHeal groups compared to the vehicle. Despite a heterogeneous distribution of values in the M1 profile, no statistical differences were detected (Fig. 9).
NeuroHeal and NeuroBoost shift the immune profile to anti-inflammatory at the early stage of response in CD4 T cells
When evaluating the involvement of CD4⁺ T lymphocytes, we observed that both treatments resulted in a reduction in the percentage of Th1 lymphocytes, which exhibit a pro-inflammatory profile, after 7 days. The Th1 cell percentage was 10 % in the NeuroBoost group and 13 % in the NeuroHeal group compared to 29 % in the vehicle group (VE: 29.46 ± 8.07 vs. NB: 10.00 ± 1.16, p = 0.003; VE: 29.46 ± 8.07 vs. NH: 13.76 ± 0.90, p = 0.02) (Fig. 10**).**Fig. 10. Characterization of CD4⁺ T Lymphocytes After spinal cord injury. Representative dot plots illustrate the flow cytometry analysis used for the phenotyping of CD4⁺ T lymphocytes (CD4⁺CD3⁺), Th1 lymphocytes (CD4⁺IFN-γ⁺), and Th2 lymphocytes (CD4⁺IL-4⁺) following treatment with vehicle, NeuroBoost, and NeuroHeal for 7 days (A), 14 days (B) and 28 days after ventral root injury (C). The percentages of Th1 and Th2 lymphocytes in the different experimental groups in respective timepoints (D-F). Two-way ANOVA - Tukey's multiple comparisons test, p < 0.05 (*).Fig. 10
After 14 days of treatment, no significant differences were observed between the evaluated groups. However, at 28 days, animals treated with the NeuroHeal combination showed an increased percentage of Th1 lymphocytes compared to the vehicle group, with 4 % in the vehicle group versus 34 % in the NeuroHeal group (4.75 ± 1.30 vs. 34.16 ± 13.33, p = 0.04). No significant differences were observed in the NeuroBoost-treated group at this time point (Fig. 10). Th2 lymphocytes appeared uniformly across treatments, with no statistically significant differences in comparison to the vehicle group at any of the three evaluated timepoints (Fig. 10)
To confirm the data found at flow cytometry data, we evaluated the expression of RNAm for each pro- and anti-inflammatory cytokine. We observed following injury, all experimental groups exhibited increased expression of the TNF-α gene compared to the control group (Control: 1.05 ± 0.16 vs. VE: 10.40 ± 1.27, p = 0.02; Control: 1.05 ± 0.16 vs. NB: 8.88 ± 2.59, p = 0.04; Control: 1.05 ± 0.16 vs. NH: 16.68 ± 2.11, p = 0.0002). In addition, TNF-α gene expression was significantly higher in the NeuroHeal group compared to the NeuroBoost group (NH: 16.68 ± 2.11 vs. NB: 8.88 ± 2.59, p = 0.04). Regarding gene expression of IFN-γ and IL-10, no significant differences were found among the evaluated groups**.** However, significant differences in IL-4 gene expression were observed in the vehicle group compared to the control group (Control: 0 ± 0 vs. VE: 1.15 ± 0.31, p = 0.02), as well as in the NeuroHeal group compared to the control (Control: 0 ± 0 vs. NH: 1.64 ± 1.32, p = 0.001).
Improvement in motor function response at the first week of treatment with NeuroBoost
In the motor functional analysis, we first analyzed the sciatic functional index, however we did not find statistical significance between the groups after 28 following-days, but we observed a qualitative improvement in animals treated with NeuroBoost, as illustrated in Fig. 11. This observation is further supported by the analysis of gait regularity (Step Sequence), which shows difference statistical analysis in NeuroBoost animals compared to vehicle-treatment group (Fig. 11).Fig. 11Functional Motor Recovery. Animals were subjected to gait analysis using the CatWalk system, starting 2 days before and following for 28 days after ventral root injury. The vehicle, NeuroBoost, and NeuroHeal treatment groups were evaluated. The Sciatic Functional Index (SFI) was used to assess motor capacity through analysis of the injured nerve (A). Gait regularity was assessed using the Step Sequence parameter (B). Representative gait patterns of animals treated with vehicle, NeuroBoost, and NeuroHeal are shown in (C). Statistical significance between the vehicle and NeuroBoost groups is indicated by * (p < 0.05). Two-way ANOVA - Tukey's multiple comparisons test.Fig. 11
Regarding the paw pressure on the glass platform, we evaluated two parameters, first MaxContact MaxIntensity, showed no statistically significant differences were observed between animals treated with NeuroBoost or NeuroHeal compared to the vehicle group (Fig. 12). However, when this analysis was separated into two independent parameters, a statistically significant difference after day 16 when a substantial improvement was observed in both treatment groups (Fig. 12). In the NeuroBoost group, this improvement persisted through days 25 and 26 post-injury. In the NeuroHeal group, the difference extended from day 21 to day 26 of treatment (Fig. 12). These findings reinforce the conclusion that both treatment combinations promoted superior motor recovery compared to the vehicle group.Fig. 12Functional Recovery. Animals were subjected to gait analysis using the CatWalk system, starting 2 days before and following for 28 days after ventral root injury. The vehicle, NeuroBoost, and NeuroHeal treatment groups were evaluated. Paw pressure on the platform was assessed through MaxContact MaxIntensity (B), Max Contact (C), and Max Intensity (D) analyses. Statistical significance between the vehicle and NeuroBoost groups is indicated by * (p < 0.05), and between the vehicle and NeuroHeal groups by # (p < 0.05). Two-way ANOVA - Tukey's multiple comparisons test.Fig. 12
Next, we evaluated the MaxIntensity parameter, only the group treated with the NeuroBoost combination exhibited a statistically significant difference compared to the vehicle group on day 6 (Fig. 12). This improvement was also evident in the gait analysis (Step Sequence), where animals treated with NeuroBoost demonstrated greater regularity in paw placement on the platform on days 5 and 6 post-injury compared to the vehicle group (Fig. 11).
In general, these set of analyses showed that treatment with both combinations resulted in an apparent improvement in the sciatic functional index. However, this improvement did not reach statistical significance when compared to the control group (Fig. 11, Fig. 12). The supplementary Table 1 contains a summary of the experiments and statistical analyses for each technique and survival time post-injury used throughout the study (Table S1).
Discussion
Nerve root injuries, often caused after motor vehicles accidents, falls, or violence, lead to extensive motoneuron degeneration. Such trauma disrupts the central-peripheral nervous system interface, contributing to a breakdown in neural homeostasis. In this context, nerve root compression, which may also occur in conditions such as herniated discs or spinal tumors, may be used as a valuable experimental model for the development and evaluation of novel therapeutic strategies.
Previous studies have shown that dimethyl fumarate (DMF) treatment following ventral root crush injury exhibits both neuroprotective and immunomodulatory properties. In mice model, 90 mg/kg dose resulted in 88 % motoneuron survival at 4 weeks, which remained at 77 % after 8 weeks (Carvalho et al., 2020). Additionally, in a model of ventral root avulsion in rats, DMF at a dose of 12 mg/kg preserved approximately 70 % of motoneurons, with other doses also showing satisfactory survival rates ranging from 52 % to 54 % (Kempe et al., 2020).
The neuroprotective effects of DMF have been extensively documented in the literature. That mechanism of action involves the activation of the Nrf2–ERK1/2 MAPK pathway, which leads to the upregulation of antioxidant enzymes such as heme oxygenase-1 (HO-1), glutathione-S-transferase, superoxide dismutase, and NAD(P)H quinone oxidoreductase-1 across various cell types (Lin et al., 2011, Linker et al., 2011, Scannevin et al., 2012, Schmidt and Dringen, 2010, Suneetha and Raja Rajeswari, 2016). Notably, DMF is routinely used for the treatment of multiple sclerosis, and in experimental models, it has been shown to reduce disease severity, particularly due to its neuroprotective mechanisms (Linker et al., 2011).
Considering antioxidant-based strategies, 4-hydroxy-TEMPO (Tempol), an antioxidant used for psoriasis model treatment, has shown promising results in a ventral root crush injury model in rats, where it reduced glial reactivity and preserved synapses from excitotoxicity (Spejo et al., 2019). Additionally, following nerve transection in neonatal rats, Tempol treatment resulted in motoneuron preservation within hours after administration, demonstrating neuroprotective potential in this model (Chiarotto et al., 2014).
Encouraged by these promising findings, we evaluated for the first time the combined administration of DMF and Tempol, named NeuroBoost, in a mouse model of ventral root crush injury. To compare this novel combination with other therapeutic options, we also investigated the neuroprotective and immunomodulatory effects of NeuroHeal, a previously established combination. NeuroHeal has been reported to promote neuroprotection and functional recovery following spinal cord injury. Among several tested drug combinations, acamprosate and ribavirin showed the best outcomes in terms of neuroprotection and functional recovery in a rat model of ventral root avulsion (Romeo-Guitart et al., 2018b).
Another study using a dorsal root contusion injury model demonstrated the neuroprotective effects of Tempol, which were associated with reduced neuronal hyperexcitability and attenuation of neuropathic pain (Quan et al., 2013). Given its function as a free radical scavenger, Tempol exhibits potential neuroprotective properties; however, in the ventral root crush injury model, Tempol alone did not show neuroprotective effects. In contrast, when combined with DMF, a significant neuroprotective outcome was observed following 14 days of treatment.
Treatment with the NeuroHeal combination also presents significant neuroprotective results, with approximately 80 % motoneuron survival detected 7 days after administration. These findings are consistent with previous studies by Romeo-Guitart et al. (2018), which reported improved neuronal survival after 21 days of NeuroHeal treatment, with 50 % of motoneurons preserved following ventral root avulsion. Therefore, even in less severe injury models, NeuroHeal appears to exert a robust neuroprotective effect on neurons (Romeo-Guitart et al., 2018a).
Moreover, in a peripheral nerve injury model, activation of the SIRT1 pathway was shown to be essential for the neuroprotective function of the combination. Similarly, in a muscle atrophy model, the treatment also demonstrated neuroprotective properties (Romeo-Guitart et al., 2018b, Marmolejo-Martínez-Artesero et al., 2020). The SIRT1 signaling pathway regulates multiple mechanisms involved in body homeostasis and has been associated with the modulation of histone 3, p53, and NFkB, displaying neuroprotective features in neural injury and neurodegenerative conditions (Romeo-Guitart et al., 2018b). Thus, the ability of NeuroHeal to activate multiple signaling cascades, particularly the overexpression of SIRT1, is likely critical for its therapeutic effects following spinal cord injury (Romeo-Guitart et al., 2018b).
Glutamate regulation within the nervous system is primarily mediated by its storage in presynaptic vesicles, particularly through the action of vesicular glutamate transporters (VGLUTs). Enhanced glutamatergic signaling, often due to an increase in these vesicles, has been implicated in several key mechanisms following nerve injury (Ghanbari et al., 2014, Inquimbert et al., 2012, Latremoliere and Woolf, 2009, Ni et al., 1994, Schäfer et al., 2002, Takamori et al., 2001). Given the importance of maintaining a balance between excitatory and inhibitory synaptic vesicles, we evaluated VGLUT-1 expression following ventral root crush injury. Our results revealed that treatment with the NeuroBoost and NeuroHeal combinations led to an upregulation of VGLUT-1, suggesting increased preservation of primary afferent inputs, corresponding mostly to proprioceptive information from the dorsal root ganglia, what may be considered positive for the sensorimotor integration.
Additionally, treatment with both NeuroBoost and NeuroHeal combinations resulted in significant modulation of astroglial reactivity, as evidenced by a reduction in astrocyte activation at 7- and 28-days post-treatment. This immunomodulatory potential has been previously demonstrated in studies using DMF in models of ventral root avulsion (Kempe et al., 2020) and ventral root crush injury (Carvalho et al., 2020), as well as in animals treated with Tempol following ventral root crush (Spejo et al., 2019).
Regarding NeuroHeal, the findings of the present study are consistent with previous reports, which astroglial reactivity was also reduced after administration of this combination in a rat model of ventral root avulsion (Romeo-Guitart et al., 2018a). The modulation or regulation of astrocyte activation contributes to the creation of a microenvironment in which the neurotoxic effects of a pro-inflammatory response are attenuated at the injury site. This, in turn, supports more efficient neural regeneration, enhances neuronal survival, minimizes tissue damage, promotes axonal growth, and facilitates remyelination by oligodendrocytes. Additionally, this process helps to limit the spread of injury to adjacent areas (Liddelow et al., 2017, Gaudet and Fonken, 2018).
Similarly, microglial activation, often abundant at inflammatory sites, can be detrimental to the regenerative process when chronically changes toward a pro-inflammatory phenotype, as it leads to the release of pro-inflammatory cytokines that exacerbate neuronal death and recruit additional inflammatory cells to the lesion site (Chen et al., 2017, Courtine and Sofroniew, 2019, Dietz and Fouad, 2014). It is important to emphasize the interplay between glial cells, as demonstrated by Liddelow and colleagues, who showed that microglia can induce a neurotoxic A1 astrocyte phenotype, which is characterized by the secretion of cytokines such as TNF and IL-1β (Liddelow et al., 2017, Assinck et al., 2017).
The search for immunomodulatory therapies aims to counteract this detrimental environment by promoting a shift toward a pro-regenerative response, thereby supporting neuronal survival and maintenance after injury. In this context, treatment with NeuroBoost after ventral root crush injury showed considerable potential, as it reduced microglial reactivity at 28 days, corroborating previous findings where DMF treatment decreased Iba-1 expression, a classical marker of microglia (Carvalho et al., 2020). Furthermore, Kempe et al. (2020) reported that even in severe lesions such as root avulsion, DMF exhibited modulatory effects by reducing microglial activation post-treatment (Kempe et al., 2020). Another study demonstrated that DMF administration in a diabetes model reduced the expression of pro-inflammatory proteins in microglia while increasing the expression of anti-inflammatory markers, thereby shifting the microglial profile toward an M2 phenotype (Lee et al., 2021).
Spejo et al. (2019) observed that treatment with Tempol significantly reduced microglial reactivity in animals subjected to ventral root crush injury, as evidenced by decreased Iba-1 expression in the spinal cord. Although Tempol did not show significant effects on neuronal survival, its role in glial modulation was of great importance (Spejo et al., 2019).
At the end of the 28-day treatment period, no significant differences in Iba-1 labeling were observed in animals treated with NeuroHeal. However, significant differences were found when comparing animals treated with NeuroHeal versus those treated with NeuroBoost, suggesting a potential reduction in reactivity after a longer treatment duration. This trend has been previously reported, where NeuroHeal administration led to reduced Iba-1 expression after 21 days of treatment in rats subjected to root avulsion (Romeo-Guitart et al., 2018a). Given that the present study employed a different injury model, a similar response may occur at different time points.
In addition to molecular changes, glial cells also undergo morphological alterations in response to activation or polarization, especially in the shape and characteristics of their projections, which are typically in contact with axotomized motoneurons. It is well established that the number of morphologically activated microglia increases after injury (Cullheim and Thams, 2007, Oliveira and Langone, 2000, Aldskogius et al., 1999, Novikov et al., 1997, Svensson et al., 1993).
Further, NeuroBoost treatment resulted in an increase in non-reactive (type II) microglia at 7 days, and a decrease in type V reactive microglia compared to the vehicle group. This observation correlates with the previous findings from immunolabeling that indicated lower microglial reactivity following treatment with the drug combination. These results were also evident at 14 and 28 days, demonstrating a higher proportion of vigilant/non-reactive microglia at the lesion site. The percentage of microglia with a vigilant/non-reactive profile at the lesion site was also evaluated. Regarding treatment with the NeuroHeal combination, we observed an increase in the non-reactive/vigilant profile and a decrease in the reactive profile after 7 days of treatment. This trend was also observed at 14 days. Although we did not find statistically significant differences through immunofluorescence quantification, morphological analysis of the microglia suggests that the treatment promoted a less toxic profile following injury, which may contribute positively to the regenerative environment.
In addition to the morphological analysis, we conducted a specific evaluation of glial and immune cell subpopulations using flow cytometry. We observed that NeuroBoost treatment led to an increase in both A1 and A2 astrocyte subtypes after 7 days of treatment compared to the vehicle, with no significant differences observed at 14 days. However, at 28 days, we noted an increase only in the pro-inflammatory A1 subtype. These findings suggest that the treatment initially induced a balanced glial response, likely aimed at limiting tissue damage during the acute phase. However, this response shifted toward a pro-inflammatory state, possibly due to the severity of the injury. Nevertheless, this early dual A1/A2 activation may have contributed to greater neuronal and synaptic preservation, which could ultimately result in improved motor coordination and control in the long term.
Similar results were observed with the NeuroHeal combination, except for a decrease in the A1 astrocyte profile after 14 days of treatment compared to the vehicle. The presence of pro-inflammatory cytokines at the lesion site can impair the regenerative process, as these cytokines activate genes associated with synaptic damage and neuronal death (Liddelow & Barres, 2017; Liddelow et al., 2017). However, when evaluating the expression of anti-inflammatory cytokines in A2 astrocytes, we found that NeuroHeal treatment increased this co-expression compared to the other groups. This may indicate a more pro-regenerative astrocytic response induced by the combination treatment. Recently, our group demonstrated that DMF and Tempol individually acts in different pathways after root crush, special Tempol decrease A1 astrocytes in the early stages of the injury, first and second week treatment (Balzani et al., 2025). These results corroborate with our finds here after the treatment with the combination NeuroBoost.
A2 astrocytes are known to support tissue repair and regeneration (Liddelow and Barres, 2017). Therefore, many therapeutic strategies aim to shift the astrocytic phenotype toward this protective state following spinal cord injury. Our findings suggest that treatment with the NeuroHeal combination may promote recovery and regeneration after injury. In addition to the role of astrocytes at the lesion site, studies show that both macrophages and microglia actively participate in the inflammatory process, producing pro-inflammatory cytokines such as TNF-α, IL-6, and IFN-γ, which inhibit repair and the return to homeostasis (Ben-Hur et al., 2003, David et al., 2012, Ekdahl et al., 2003, Prüss et al., 2011, Serhan et al., 2007). Moreover, once microglia/macrophages adopt a reactive phenotype, therapeutic strategies are required to return them to a resting state, as chronic gliosis is highly detrimental, even to areas not initially affected by the injury. This phenomenon leads to the expansion of the lesion epicenter, damaging healthy tissue and resulting in significant functional loss (Wu et al., 2014, Zhao et al., 2007).
We observed that animals treated with NeuroBoost showed a reduction in the M1 microglial profile at 7 days and a decrease in the percentage of M2 microglia at 28 days, which corroborates previous morphological findings. A similar pattern was seen in animals treated with NeuroHeal, where a decrease in M2 microglia was observed at 28 days. However, no significant differences were found in Iba-1 immunofluorescence quantification when compared to the vehicle group. Additionally, an increase in M1 macrophages at the lesion site was noted after 14 days of treatment with the combination. Similar results were described by Balzani and colleagues (2025), where Tempol decrease both M1 and M2 microglia after 14-days treatment, and followed until 28 days after the injury (Balzani et al., 2025).
Regarding the percentage of T lymphocytes (Th1 and Th2), both treatments resulted in a reduction in Th1 cells, which are associated with a pro-inflammatory profile, suggesting a shift toward an anti-inflammatory response. Notably, DMF has been previously described in the literature as capable of promoting a Th2 response, reducing the expression of pro-inflammatory cytokines (Ghoreschi et al., 2011). Furthermore, DMF induces the depletion of memory B and T cells, leading to a state of immune tolerance (Fleischer et al., 2018, Longbrake et al., 2016). Another study showed that DMF increases the expression of CD56^high^ NK cells and TNF-α-producing CD8 T cells, contributing to its effectiveness in the treatment of MS (Longbrake et al., 2016, Gross et al., 2016, Li et al., 2017, Medina et al., 2018, Wu et al., 2017).
Overall, the NeuroBoost combination demonstrated a more classical immune modulation profile, particularly in terms of lymphocyte response, when compared to the individual drugs. In the ventral root crush scenario were demonstrated that DMF decrease Th1 response after 28 days treatment, and Tempol acts in the early stage of the inflammatory process, after 7-days treatment (Balzani et al., 2025).
Bombeiro et al. (2020) indicated that, following a nerve crush injury, treatment with DMF results in greater recruitment of macrophages to the lesion site immediately after injury, followed by a decrease in both M1 and M2 macrophage subpopulations after 14 days of treatment (Bombeiro et al., 2020). These results, similarly, to our findings, suggest that the macrophage/microglial response is predominantly active during the initial phase of the inflammatory process. At this stage, a balance between pro-inflammatory and anti-inflammatory responses is observed, indicating a repair and inflammation-resolution phenotype.
However, by 28 days post-injury, macrophages/microglia are still present at the lesion site but now exhibit a pro-inflammatory profile. These results can be associated with our observations, even though there are differences between the inflammatory response in the nerve and in the spinal cord. Nonetheless, it is suggested that the inflammatory state observed in the present study might be associated with the inflammation-resolution process, potentially beneficial to the microenvironment (Kigerl et al., 2009), as the disruption of the blood–brain barrier initiates a chemotactic process, attracting immune cells to the injury site. This phenomenon leads to a marked influx of macrophages and lymphocytes. As previously discussed, macrophages play a critical role in the clearance of cellular debris and support the inflammatory response (Kleinschnitz et al., 2013).
In addition to the points previously discussed, when we examined the relative gene expression data for pro- and anti-inflammatory cytokines, we observed that the balanced response identified in astrocytes and microglia by flow cytometry was also reflected in the gene expression profile. There was an increase in the relative expression of TNF-α and IL-4 after 7 days of treatment with both drug combinations, a finding that is consistent with previous data reported by Carvalho et al., 2020 following DMF treatment (Carvalho et al., 2020). Taken together, our observations reinforce findings from studies in patients with multiple sclerosis, where DMF administration led to a reduction in CD4 + and CD8 + T lymphocytes, as well as B cells. Moreover, a decrease in Th1 and Th17 lymphocytes was observed, while Th2 cells increased (Wu et al., 2017). This suggests that the effect of DMF is translatable to this experimental model of motor root injury, where it induces a shift toward an anti-inflammatory profile. The literature also supports that DMF reduces oxidative stress and modulates immune responses through activation of the Nrf2 pathway (Linker et al., 2011).
In the evaluation of functional recovery, we observed that treatment with NeuroBoost led to improvements in the functional recovery as early as 5 days post-treatment. Particularly in the analysis of gait regularity through the Step Sequence parameter, animals treated with NeuroBoost exhibited greater regularity compared to vehicle-treated animals. Additionally, increased paw pressure was observed through the Max Contact and Max Intensity parameters, indicating improvement beginning on the 6th day of treatment with the combination. Regarding animals treated with NeuroHeal, significant improvements were observed only in the paw pressure analysis via Max Contact, which became evident from the 16th day of treatment.
Although no statistically significant differences were found in the Sciatic Functional Index (SFI), it is possible to discern a notable improvement in the recovery of animals in both treatment groups when compared to vehicle controls. The combined analysis of these data suggests an enhanced in the capacity of the affected limb and progressive improvement in gait, although paw positioning, plantar load distribution, and motor function have not yet reached an optimal recovery state. Overall, the intersection of these factors, especially evident in the early phases post-injury, suggests the potential for normalization of gait biomechanics over time.
Correlation of these findings with previous studies reveals that dimethyl fumarate (DMF) exerts a beneficial effect on functional recovery. Prior research using root crush and avulsion models demonstrated that DMF significantly improved functional outcomes after four weeks of treatment (Carvalho et al., 2020, Kempe et al., 2020). Regarding treatment with NeuroHeal, we did not observe significant differences in motor recovery compared to the vehicle group. This contrasts with findings from a study conducted by Romeo-Guitart et al. (2018), which reported significant functional recovery following treatment with the combination (Romeo-Guitart et al., 2018a). However, it is important to consider that their study was conducted in rats, and discrepancies may be attributed to differences in recovery between rats and mice (Benus et al., 1987).
Overall, the results of the present study indicate that the NeuroBoost combination led to greater neuronal preservation, followed by notable neuronal survival at 14 days of treatment, and a reduction in neuronal excitability, thereby preventing neuronal death. These findings correlate with the functional recovery data, where animals treated with the combination showed functional improvement during the first week of treatment. A similar observation was made in animals treated with NeuroHeal; however, despite earlier neuronal preservation, no significant functional recovery was observed throughout the treatment period. Only at the end of treatment were minor improvements in gait detected.
Materials and methods
Experimental design
Eight-week-old female C57BL/6 J mice were utilized in this study, in accordance with the guidelines from the National Council for Animal Experimentation Control (CONCEA – Brazil). The experimental protocol was approved by the Institutional Ethics Committee on the Use of Animals (CEUA/IB/UNICAMP), under protocol numbers 5327–1/2019 and 5727–1(A)/2023. Throughout the experimental period, the animals were housed in the animal facility under a 12-hour light-dark cycle, with unrestricted access to water and food.
The mice underwent a ventral lumbar root crush at the L4, L5, and L6 spinal segments. Anesthesia was induced using a combination of ketamine (100 mg/kg, Fort Dodge, USA) and xylazine (20 mg/kg, König, Argentina), adjusted according to body weight. The animals were maintained under 2 % isoflurane during the surgical procedure. A midline incision was made, providing access to the spinal cord and allowing unilateral root crush of the L4, L5, and L6 spinal segments.
Post-surgery, the incisions were sutured, and the animals received oral Tramadol hydrochloride (Germed Farmacêutica Ltda, São Paulo, Brazil) for three days at a dose of 5 mg/kg. Subsequently, the animals were divided into three groups: vehicle, NeuroBoost, and NeuroHeal, and were monitored at 7, 14, and 28 days post-procedure.
NeuroBoost treatment consisted of a combination of dimethyl fumarate (90 mg/kg, Sigma-Aldrich, USA) and 4-hydroxy-TEMPO (Tempol) (50 mg/kg, Sigma-Aldrich, USA), administered orally daily during the treatment period. NeuroHeal treatment included Acamprosate (100 mg/kg, Cayman Chemical Company, USA) and Ribavirin (70 mg/kg, prepared formulation), also administered orally daily. In the vehicle group, animals received daily oral administration of methylcellulose during the specified period.
Functional analysis
Animals were evaluated 2 days before the surgical procedure and 28 days daily after the procedure, starting at 3rd day post-op. The platform's speed and the pressure exerted were pre-determined to accurately record the runs. The duration of each run was established between 0.5 and 5.0 s, with an allowable variation of up to 40 %. The functional assessment system includes a fluorescent light within the transparent floor, which highlights the contact points of the paw prints based on the pressure exerted by the animals' paws during locomotion. The signal strength varies in accordance with the pressure applied by the animal's paw. They were subjective three runs at catwalk when was monitored by Fujinon DF6HA-1B camera (Cosimar 8,5 mm) utilizing following parameters: gait (dB) 20,00, intensity green light 0,10, voltage 16,0 and, red light voltage 17,7, evaluated by Catwalk XT 10.5 (Noldus Inc.).
To evaluate motor coordination parameters, we used the distance between the first and the fifth toe (toe spread, TS), and the distance between the third toe and heel (print legth, PL) to express the sciatic functional index (SFI) according with Inserra (Inserra et al., 1998). Also, to evaluate the regularity index (Step Sequence), we used values from the four paws of animals running at the catwalk after three runs in relation to normal values before the surgical procedure.
To analyze the intensity and pressure exerted by the animal on the platform, the time in seconds was measured from the start of the run to the point when the widest region of the paw print made maximum contact with the platform glass (Max Contact), corresponding to the duration of the paw's contact with the glass. Additionally, to assess intensity (Max Intensity), the maximum intensity of the paw print on the platform glass during the run was recorded.
Perfusion and specimen preparation
Animals were euthanasia after 7, 14 and 28 days post-op following overdose of ketamine (100 mg/kg, Fort Dodge, USA) and xylazine (20 mg/kg, König, Argentina), and were perfused with saline (NaCl 0.9 % - PB 0.1 M, pH 7.38) and fix solution (paraformaldehyde 4 % - PB - 0.1 M, pH 7.38). Lumbar intumescence was dissected out and incubated with fix solution during 24 h, following tissue preservation with increasing sucrose solutions (10 %, 20 % and 30 % - PB 0.1 M) for 24 h each concentration solution. Specimen was included in Tissue-Tek (Miles Inc., USA) and frozen at −35ºC, and cross sections with 12um thickness was made at cryostat (HM525, Microm), and slides were kept at −20 ºC.
Motorneuron survival – Nissl stain
Alternate slides were stained with toluidine blue (0.05 % in distilled water), dehydrated and diaphanized. The slides were analyzed at light microscope (DM550B, Leica Microsystems), and motorneuron presents at lamina IX of Rexed were counted in both sides of spinal cord according with Abercrombie formula: N = nt/(t + d), where N is the corrected number of neurons, n is the number of cells counted, t is the thickness of each slide, and d is the mean neuron diameter.
Cell count from ipsi/contralateral sides of spinal cord from L4, L5 and L6 medullar segments was calculated and made a the ipsi/contralateral ratio.
Immunofluorescence
Slides were climatized and selected from each medulla segment (L4, L5 and L6) to each antibody. They were washed with PB 0.01 M and subsequently incubated with blockage solution (BSA 3 % + PB 0.1 M) for 1 h. Then, they were stained with primary antibodies (BSA 1.5 % + Tween 0.2 % + PB 0.1 M) (Table 2) and incubated for 4 h.Table 2. Primary and secondary antibodies used to analyze glial reaction and synapse coverage.Table 2AntibodyManufacturerHostLot. NumberDiluitionIba-1WakoRabbit019–197411:750GFAPAbcamRabbitAb72601:750SynaptophysinNovus BiologicalsRabbitNBP2–251701:1000VGLUT-1Synaptic SystemsRabbit1353031:1000NeuNMilliporeMouseMAB3771:1000Alexa488JacksonRabbit711–545–1521:250Alexa 594JacksonMouse715–585–1501:250
After, slides were washed and stained with secondary antibody for 45 min and then washed 3x with PB 0.1 M and mounted with glycerol/PB.
The slides were analyzed at microscope (Leica DM 5500B fluorescence), and the lamina IX of Rexed was evaluated in both sides of the spinal cord using cyanine 2 (Cy2, 492 nm) and cyanine 3 (Cy3, 550 nm). Imagens were taken in 20x and quantified by the integrated density of pixels using Image J software (1.33 u version, National Institute of Health, USA) following previously articles (Cartarozzi et al., 2019, Oliveira et al., 2004).
Classification of microglial cells
Microglial morphology was determinate using Image J software (NIH, USA) and cells were classified into 5 types: type I: cells with few cellular processes (two or less); type II: cells with 3 or 5 short branches; type III: cells with more than 5 long branches with small cell body; type IV: cells with large cell body and retracted processes; and type V: cells with ameboid soma and small processes. They were classified as surveillant microglia (type I and type II) and activated microglia (type III, IV and V) (Diz-Chaves et al., 2012, Lopez-Rodriguez et al., 2015).
Flow cytometry
Followed the procedure validated by Coser, et al., 2024 (Coser et al., 2024). Animals were euthanasia after 7-, 14- and 28-days post-op following overdose of ketamine (100 mg/kg, Fort Dodge, USA) and xylazine (20 mg/kg, König, Argentina) and were perfused with saline (NaCl 0.9 % - PB 0.1 M, pH 7.38) and lumbar intumescence was dissected out. Lumbar intumescence was removed utilizing surgery microscopy (DFV Vasconcelos, Brazil), and ipsilateral side was isolated. Tissue was dissociated in 150uL of PBS, and spinal cord was mechanically dissociated by passing trought 140μm and 70μm meshes. Cells were centrifuged at 400x g for 10 min at 4°C, and then the pellet was resuspended at different Percoll concentrations, initially 70 %, followed 50 %, 37 % and 10 % (Sigma, USA). For cell isolation, tubes were centrifuge at 400x g for 30 min at 4°C (acceleration 8 and deceleration 0). Then, the 10/37, 37/50 and 50/70 fractions were collected and centrifuge again for pellet obtention. Using 48-well plate, cells were resuspended in 1 mL of DMEM supplemented with 10 % FBS and 1 % penicillin/streptomycin (1 mg/1 g Vitrocell, Brazil), and stimulated with PMA (50 ng/mL), ionomycin (250 ng/mL) and brefeldin A (1ug/mL), incubated for 3 h at 37°C with 5 % CO_2_.
Next, plate was centrifuge at 400x g for 10 min at 4°C, and cells were resuspended with the surface antibodies diluted in PBS-BSA-A (PBS – 0,1 % Bovine Serum Albumin – 0,5 % Sodium Azide) (Table 3), and incubate for 30 min at 4°C. Then, cells were washed with PBS-BSA-A, and fixed for with commercial buffer (kit True Nuclear Transcription Buffer Set, Biolegend, USA).Table 3. Antibodies used for characterization of glial and immune cells by flow cytometry.Table 3LayerFITCPEPE-Cy5PE-Cy7APCAPC-Cy710/37GFAPIL-10TNF-a-IL-4-37/50CD45IL-10TNF-aCD206CD68CD11b50/70CD4--IFN-gIL-4CD3
Cells were washed again, and permeabilized with commercial buffer (Perm Buffer – True Nuclear Transcription Buffer Set, Biolegend, USA) and incubate overnight at 4°C with the intracellular antibodies (Table 3). At the second-day assay, cells were washed with Perm Buffer and fixed. Cells were acquired at NovoCyte flow cytometry (ACEA biosciences, USA), and the data were analyzed using software NovoExpress (version 1.6.2), following minimum 10.000 events per sample. Our gate strategy we use unstained cells to gate the population of interest and their respective subpopulations, avoiding autofluorescence.
qRT-PCR
Animals were euthanized following the same process from flow cytometry section, after lumbar intumescence was dissected out, the specimens were frozen in liquid nitrogen, and then stored at −80°C until process.
RNA extraction was performed using QIAzol lysis reagents (Qiagen, Germany), following cDNA syntesis using 1,5μg of High-Capacity cDNA Reverse Transcription kit (Applied Biosystems), according to manufacturer’s instructions. Then, we used for Real Time PCR, utilizing Taqman Gene Expression Master Mix (2x) (Life Tecnologies) to analysis gene expression (Table 4).Table 4. Genes of interest used for TaqMan assay in qRT-PCR analysis.Table 4GeneAssay ID**GAPDHMm99999915_g1HPRT1Mm01545399_m1TNFαMm00443258_m1IFNγMm01168134_m1IL-10Mm01288386_m1IL-4Mm00445259_m1
Conclusion
We conclude that both pharmacological combinations demonstrated relevant neuroprotective and immunomodulatory effects following ventral spinal root crush injury. Treatment with the NB combination resulted in improvement of motoneuron survival at 14 days post-treatment compared to the vehicle group, while animals treated with NH showed increased motoneuron survival as early as 7 days post-treatment. Both NB and NH modulated the astroglial response; however, only NB significantly reduced microglial reactivity compared to the vehicle-treated group. Although no significant differences in synaptic coverage were observed between groups, both combinations led to increased labeling of excitatory vesicles (VGLUT-1) relative to vehicle controls. Despite the reduction in astroglial reactivity, both treatments were associated with a higher percentage of A1 astrocytic phenotype during the post-injury response. Moreover, NB and NH induced a shift toward a non-reactive/surveillant microglial profile, with structural features consistent with type I and II microglia. Both treatments also led to a reduction in the pro-inflammatory Th1 lymphocyte profile in the early stages following injury. Gene expression analysis revealed a balanced expression of pro- and anti-inflammatory cytokines after 7 days of treatment, indicating an immunomodulatory effect. Functional recovery data demonstrated therapeutic potential of these treatments, with NB showing superior performance, as evidenced by significant improvements in plantar print pressure and gait regularity (Max Contact/Max Intensity; Step Sequence). Together, our results suggest that both NeuroBoost and NeuroHeal present neuroprotective and immunomodulatory potential in the context of proximal nerve injury at the CNS/PNS interface. Further studies are needed to determine the specific conditions under which each strategy may be most effective, supporting their translational application in clinical settings.
CRediT authorship contribution statement
Júlia Lombardi: Writing – original draft, Investigation, Formal analysis, Data curation. Luciana Politti Cartarozzi: Writing – original draft, Methodology, Data curation, Conceptualization. Lilian de Oliveira Coser: Writing – original draft, Methodology, Investigation, Conceptualization. Maria Fernanda Vannucci: Writing – original draft, Formal analysis, Data curation. Alexandre Leite Rodrigues de Oliveira: Writing – original draft, Supervision, Methodology, Investigation, Funding acquisition, Formal analysis, Conceptualization.
Funding
This research was funded by CNPq (Nacional Council for Scientific and Technological Development) 140027/2020–3, 303050/2021–7 and FAPESP (Sao Paulo Research Foundation) 2018/05006–0, 2018/17554–2, 2023/02615–4, 2023/16415–7).
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Mazzer P.Y.Barbieri C.H.Mazzer N.Fazan V.P.Morphologic and morphometric evaluation of experimental acute crush injuries of the sciatic nerve of rats J. Neurosci. Methods 173200824925810.1016/j.jneumeth.2008.06.01918644327 · doi ↗ · pubmed ↗
- 2Kobayashi S.Yoshizawa H.Yamada S.Pathology of lumbar nerve root compression. Part 2: morphological and immunohistochemical changes of dorsal root ganglion J. Orthop. Res 22200418018810.1016/s 0736-0266(03)00132-314656678 · doi ↗ · pubmed ↗
- 3Anderson K.D.Targeting recovery: priorities of the spinal cord-injured population J. Neurotrauma 2120041371138310.1089/neu.2004.21.137115672628 · doi ↗ · pubmed ↗
- 4Thuret S.Moon L.D.Gage F.H.Therapeutic interventions after spinal cord injury Nat. Rev. Neurosci.7200662864310.1038/nrn 195516858391 · doi ↗ · pubmed ↗
- 5Spejo A.B.Carvalho J.L.Goes A.M.Oliveira A.L.Neuroprotective effects of mesenchymal stem cells on spinal motoneurons following ventral root axotomy: synapse stability and axonal regeneration Neuroscience 250201371573210.1016/j.neuroscience.2013.07.04323896572 · doi ↗ · pubmed ↗
- 6Aldskogius H.Mechanisms and consequences of microglial responses to peripheral axotomy Front Biosci. (Sch. Ed. )3201185786810.2741/19221622237 · doi ↗ · pubmed ↗
- 7Cullheim S.Thams S.The microglial networks of the brain and their role in neuronal network plasticity after lesion Brain Res Rev.552007899610.1016/j.brainresrev.2007.03.01217509690 · doi ↗ · pubmed ↗
- 8Oliveira A.L.Langone F.GM-1 ganglioside treatment reduces motoneuron death after ventral root avulsion in adult rats Neurosci. Lett.293200013113410.1016/s 0304-3940(00)01506-811027851 · doi ↗ · pubmed ↗