Study On The Neuroprotective Effects Of Molecular Hydrogen During The Secondary Spinal Cord Injury In Rats

BackgroundSpinal cord injury (SCI) causes high rate of disability, and seriously affects thequality of life, also, it makes great economic losses to the society. The majority of spinalcord injury is caused by the strong impact, then the dislocation or fracture of vertebraeinduces the compression or contusion of spinal cord, and this is the primary spinalcord injury. However, during the progress of spinal cord injury, the secondary spinal cordinjury is often more disruptive. Primary, traumatic spinal cord injury induces the releaseof reactive oxygen species (ROS), such as superoxide anion (O2), hydrogen peroxide(H2O2) and hydroxyl radical (OH), within several minutes to hours. ROS is consideredto be a crucial factor related to secondary SCI; it can induce a series of cascade reactions,which not only aggravate the dysfunction or death of neurons, but also expand the scopeof injury and inflammation. In addition, ROS is one of the main causes of astrocyteactivation. Astrogliosis leads to the excessive secretion of proteoglycans and intermediatefilaments, such as glial fibrillary acidic protein (GFAP), which leads to the formation ofthe glial scar which has been regarded as a critical factor hindering axonal repair andregeneration. These inhibitory molecules then diffuse and will cause the injury ofneighboring neurons. Activated astrocytes can release large amounts of inflammatorycytokines, such as IL-1β, IL-6and TNF-α, which not only expand the tissueinflammation, but also continuously activate itself in the form of autocrine to generatemore ROS and inflammatory factors, and cause neurotoxicity as well as a vicious cycle.Currently, there is no antioxidant drugs with efficacious neuroprotection can be usedclinically.Hydrogen has been proven to be a safe and effective, novel antioxidant. In animalmodels, inhalation of hydrogen gas or intraperitoneal injection of hydrogen-rich saline(HS) can effectively inhibit ischemia-reperfusion-induced oxidative stress in brain, liver, heart, intestine and other tissues, as well as the inflammation induced by intestinaltransplantation or acute pancreatitis. Hydrogen products have fewer adverse effects thanmost known antioxidants, because of their ability to selectively scavenge detrimentalROS, such as HO and peroxynitrite (ONOO), without disturbing metabolicoxidation–reduction reactions, and because hydrogen products are able to penetratebiomembranes and diffuse into the cytosol, mitochondria and nucleus. Moreover, ourpreliminary study has demonstrated that, after SCI, intraperitoneal injection of HS canpromote the recovery of locomotor function. However, it remains unknown what are thepotential mechanisms of this spinal cord protective effect.In this present study, we used a rat contusive SCI model (in vivo), and two inducedoxidative damage cell models (in vitro), to verify that molecular hydrogen on the onehand could suppress reactive astrogliosis along with glial scar and inflammationexpansion, on the other hand, it could inhibit the apoptosis and promote the growth ofneurons induced by oxidative injury during SCI. Therefore, it possesses the effect ofneuroprotection. The results will provide new theoretical basis and treatment strategiesfor early intervention during spinal cord injury to improve functional recovery.Methods1. The tissue inflammation, astrogliosis, and locomotor function ofthe experimental rats were evaluated after Allen’s SCI modelperformedBBB score had been evaluated before surgery. Rats were randomly divided intothree groups as follows:1) Sham-control group: sham-operated plus normal saline (NS,0.9%NaCl)(n=24), rats of this group had their dura exposed without contusive SCI;2)Allen+NS group: Rats were given Allen’s SCI surgery plus NS (n=24);3) Allen+HSgroup: SCI plus HS (n=24). Immediately after the surgery had been accomplished, NS orHS was injected (8ml/kg, i.p.), and then was administrated every12h until the animalswere sacrificed. The release of three proinflammatory cytokines (IL-1β, IL-6and TNF-α)in the spinal cord tissue and the expression of STAT3, p-STAT3and GFAP were assessedon POD3, and the level of astrocyte activation on POD7,14along with locomotorfunction (BBB score) were evaluated on POD1,7,14. 2. Purified astrocytes culture, and investigated the effect ofhydrogen-rich medium (HM) on asrogliosis induced by oxidative injuryPure astrocytes were cultured and divided into four groups:1) Control: culturedwith normal medium (NM);2) HM: cultured with hydrogen-rich medium (HM);3) NM+H2O2: cultured with NM, and treated with Feton reaction (100μM H2O2along with15μM FeCl2);4) HM+H2O2: cultured with HM containing100μM H2O2and15μM FeCl2after pretreatment of HM for2h. Treatments were carried out for12h, then theintracellular ROS and hydroxyl radical (OH) production, the level of GFAP and BrdU,as well as the release of IL-1β, IL-6and TNF-α were assessed.3. Purified spinal cord neurons culture, and investigated the effectof hydrogen-rich medium (HM) on apoptosis and growth after oxidativeinjuryPure spinal cord neurons were cultured and divided into four groups:1) Control:cultured with normal medium (NM);2) HM: cultured with hydrogen-rich medium (HM);3) NM+H2O2: cultured with NM, and treated with Feton reaction (100μM H2O2alongwith15μM FeCl2);4) HM+H2O2: cultured with HM containing100μM H2O2and15μM FeCl2after pretreatment of HM for2h. Treatments were carried out for12h, thenthe intracellular ROS and hydroxyl radical (HO) production, the number of TUNELpositive cells and caspase-3level, as well as the expression of GSK-3β and p-GSK-3βwere assessed.4. Statistical analysisStatistical analysis was performed by SPSS17.0software, and the data are expressedas mean±Standard Deviation (x±S) for each group. The differences among groups wereanalyzed by Student’s t-test and one-way ANOVA. A two-way ANOVA withrepeated-measures was used to compare matched data at multiple time points. P <0.05wasconsidered statistically significant. Results1. Hydrogen-rich saline attenuated the release of proinflammatorycytokines, suppressed the reactive astrogliosis, and promoted therecovery of hind limb locomotor function after SCIThe content of IL-1β, IL-6and TNF-α in local contusive spinal cord was notablyattenuated by hydrogen-rich saline (Allen+HS vs. Allen+NS:431.62±50.71vs.653.49±69.89, p <0.01;101.40±11.41vs.134.54±21.83, p <0.01;718.29±127.40vs.1085.892±319.73, p <0.05, respectively), and the expression of STAT3, p-STAT3and GFAP was reduced (0.40±0.03vs.0.66±0.04，p <0.01；1.09±0.09vs.1.26±0.05，p <0.05；1.03±0.05vs.1.45±0.09，p <0.01, respectively) on POD3. In addition,the expression of GFAP (measured by the fluorescence intensity) was dramaticallysuppressed on POD7and14(1952.21±88.67vs.2386.92±273.73，p <0.01；2684.07±286.01vs.7821.48±122.24，p <0.01, respectively). Moreover, rats of the HS groupexhibited higher BBB scores on POD7and14(7.33±0.52vs.5.33±0.82，p <0.01；10.83±0.75vs.7.67±0.52，p <0.01, respectively).2. Hydrogen-rich medium reduced the intracellular production ofROS and inhibited the activation of the purified astrocytes afterH2O2-induced oxidative injury12hours after adding H2O2(100μM) and FeCl2(15μM) to the medium with orwithout super-saturated hydrogen, the production of intracellular ROS (assayed by DCFintensity) and HO (selectively assayed by HFP intensity) was significantly decreased byhydrogen-rich medium pretreatment (HM+H2O2vs. NM+H2O2:456.61±132.69vs.1827.84±169.07, p <0.01;771.04±47.57vs.1257.27±210.93, p <0.01, respectively).Also, hydrogen-rich medium cultured astrocytes had a significantly lower percentage ofBrdU-positive cells (DNA proliferation) and more less expression of GFAP (49.67±2.52vs.91.67±2.52, p <0.01;2619.56±503.58vs.6275.07±233.05, p <0.01,respectively). Meanwhile, the functional activation of astrocytes was inhibited, whichwas reflected by de decrease of IL-1β, IL-6, and TNF-α secreted by them (723.99±124.58vs.917.15±74.42, p <0.01;2081.48±38.22vs.2605.61±47.32, p <0.01; 2449.78±321.75vs.3356.18±195.41, p <0.01, respectively).3. Hydrogen-rich medium reduced the intracellular production ofROS and the apoptosis of neurons after H2O2-induced oxidative injury12hours after H2O2-induced oxidative injury, the intracellular production of ROSand HO in purified neurons could remarkably reduced by hydrogen-rich medium (HM+H2O2vs. NM+H2O2:569.40±68.00vs.1834.00±181.50, p <0.01;759.80±65.93vs.1855.00±216.20, p <0.01, respectively), and the percentage of TUNEL-positive cellsalong with the caspase-3fluorescence intensity were decreased, too (31.67±3.51%vs.60.67±4.93%, p <0.01;480.60±72.98vs.1425.00±73.92, p <0.01, respectively).4. Hydrogen-rich medium promoted the phosphorylation ofGSK-3β of neurons after H2O2-induced oxidative injury12hours after H2O2-induced oxidative injury, there was no significant differenceamong groups in the total expression of GSK-3β (p>0.05), but the inactivation(phosphorylation) of GSK-3β was reduced by oxidative injury (NM+H2O2vs. NM:0.35±0.04vs.0.78±0.06, p <0.01), and hydrogen-rich medium could promote thephosphorylation of GSK-3β (HM+H2O2vs. NM+H2O2:0.54±0.08vs.0.35±0.04, p<0.01).ConclusionIn conclusion, molecular hydrogen suppresses reactive astrogliosis followed by glialscar formation after contusive SCI, and reduces the inflammation of spinal cord. Inaddition, molecular hydrogen reduces the intracellular production of ROS and especially OH in astrocytes and neurons, also, it can suppress excessive astrogliosis afterH2O2-induced oxidative injury and the secretion of proinflammatory cytokines (IL-1β,IL-6, and TNF-α) produced by active astrocytes related to oxidative injury. Furthermore,molecular hydrogen inhibits the apoptosis of neurons and increases the phosphorylationof GSK-3β to promote the growth of neurons, thus, it possesses the ability ofneuroprotection. Notably, anti-astrogliosis carried out by molecular hydrogen is anotherpotential mechanism of its neuroprotective effect. All these results support its clinical application during SCI to promote axonal regeneration and repair, as well as functionalrestoration.