Vulnerability of central neurons to secondary insults after in vitro mechanical stretch

Abstract

Mild traumatic brain injuries are of major public health significance. Neurons in such injuries often survive the primary mechanical deformation only to succumb to subsequent insults. To study mechanisms of vulnerability of injured neurons to secondary insults, we used an in vitro model of sublethal mechanical stretch. Stretch enhanced the vulnerability of the neurons to excitotoxic insults, causing nuclear irregularities, DNA fragmentation, and death suggestive of apoptosis. However, the DNA degradation was not attributable to classical (caspase mediated) or caspase-independent apoptosis. Rather, it was associated with profound stretch-induced mitochondrial dysfunction and the overproduction of reactive oxygen species (ROS). Sublethally stretched neurons produced surprisingly high levels of ROS, but these in isolation were insufficient to kill the cells. To be lethal, the ROS also needed to combine with nitric oxide (NO) to form the highly reactive species peroxynitrite. Peroxynitrite was not produced after stretch alone and arose only after combining stretch with an insult capable of stimulating NO production, such as NMDA or an NO donor. This explained the exquisite sensitivity of sublethally stretched neurons to a secondary NMDA insult. ROS scavengers and NO synthase (NOS) inhibitors prevented cell death and DNA degradation. Moreover, inhibiting neuronal NOS activation by NMDA using peptides that perturb NMDA receptor-postsynaptic density-95 interactions also reduced protein nitration and cell death, indicating that the reactive nitrogen species produced were neuronal in origin. Our data explain the mechanism of enhanced vulnerability of sublethally injured neurons to secondary excitotoxic insults and highlight the importance of secondary mechanisms to the ultimate outcome of neurons in mild neurotrauma.

Figure 5.

Stretch causes a reduction in mitochondrial membrane potential. Cultures were pretreated with 10 nm TMRM for 30 min before each experiment (see Materials and Methods). A, Effects of the indicated conditions on TMRM fluorescence at the indicated times after insult. The bars indicate background-subtracted fluorescence of cultures exposed to an insult (F_t) normalized to background-subtracted TMRM fluorescence in unstretched cells (F_o) at the indicated time points. The bars represent mean + SE of three to four cultures from n = 3 separate experiments. The asterisks indicate the difference from unstretched cultures at the indicated time point (Bonferroni t test; p < 0.05). B, Representative TMRM fluorescence images from the indicated condition. TMRM fluorescence in cultures exposed to stretch only recovered, whereas TMRM fluorescence from cultures exposed to stretch plus NMDA did not. Treatment with the mitochondrial uncoupler FCCP abolished TMRM fluorescence.

Figure 10.

Effect of uncoupling NMDAR NR2B from PSD-95 on ROS production and protein nitration. A, Schematic illustrating the approach: i, NMDARs associate with nNOS via PSD-95; ii, iii, dissociating nNOS from NMDARs using Tat fused either to the C terminus of NR2B (Tat-NR2B9c; ii) or to the first and second PDZ domains of PSD-95 (pTat-PDZ1-2; iii). B, Tat peptides and fusion proteins (pTat) used in these experiments. Inset, Representative immunoblots obtained during purification of pTat-PDZ1-2 and pTat-GK proteins. C, Visualization of intraneuronal accumulation of Tat-NR2B9c-dansyl (10 μm) but not Tat-38-48-dansyl (10 μm) 30 min after application to cortical cultures. D, Effect on DHR fluorescence of pretreating cultures with Tat-NR2B9c 30 min before the indicated insult. Cultures were simultaneously preincubated with 10 μm DHR for 30 min before the insult. Pretreatment with 50 nm TatNR2B9c reduced ROS production in all stretched (bottom) and unstretched (top) cultures treated with NMDA (30 μm or 1 mm; Boneferroni t test; p < 0.05). However, Tat-NR2B9chas no effect on ROS production by stretch alone (t₂₄ = 1.10; p = 0.284). Symbols are the means ± SE of 5-18 cultures from three separate experiments. E, Tat-NR2B9c pretreatment reduces NMDAR-mediated protein nitration. Nitrotyrosine immunostaining was performed 12 hr after the insult under the indicated conditions (representative of 3 experiments).

Figure 1.

Sublethal stretch injury renders cortical cultures vulnerable to low concentrations of NMDA. NMDA was applied for 1 hr with in 10 min of stretch. A, Effects of NMDA at the indicated concentration with or without previous stretch (130% for 1 sec). Cell death was measured at 20 hr. The asterisk indicates the difference from unstretched (t₆₁ = 13.23; p < 0.001). The bars show the mean ± SE of 27-34 cultures from three separate dissections. B, Representative phase-contrast and PI fluorescence images of unstretched and stretched cultures 20 hr after challenge, as indicated. C, High magnification of Hoechst-stained neuronal nuclei 20 hr after the indicated challenge. NMDA (1 mm or 30 μm) applied to unstretched cultures did not affect the round nuclear morphology (white arrows). However, stretch caused condensation and irregularity of nuclear morphology (open arrows). Images were obtained using identical excitation, emission, and camera gain settings (representative of 3 separate experiments).

Figure 2.

Challenging sublethally stretched neurons with low NMDA concentrations produces DNA damage. NMDA was applied for 1 hr within 10 min of stretch. A, TUNEL staining 20 hr after injury under the indicated conditions using the DAB (Ai) or FITC (Aii) methods. Nuclei in Aii were also counterstained with Hoechst. Each panel is representative of three experiments. B, Quantification of TUNEL staining at 20 hr by each method. TUNEL-positive cells were normalized to total cell number. Approximately 100-200 cells were counted per culture. Compared with unstretched controls, treatment with 1 μm staurosporine for 48 hr and stretch plus NMDA for 1 hr resulted in increased TUNEL staining (staurosporine: FITC-t₁₈ = 7.63, p < 0.0001; DAB-t₅₃ = 30.70, p < 0.0001; stretch plus NMDA: FITC-t₂₉ = 10.298, p < 0.0001; DAB-t₈₃ = 23.923, p < 0.0001). Treatment with NMDA alone (30 μm to 1 mm) or stretch alone did not result in increased TUNEL staining. The asterisks indicate the difference from unstretched control (Bonferroni t test; p < 0.05). The bars represent the mean + SE of four to eight fields in each of three cultures in each of three (FITC) or six (DAB) experiments. C, Staurosporine and stretch plus 30 μm NMDA, but not 1 mm NMDA alone or stretch alone, induced DNA laddering (arrowheads) 20 hr after insult. The data are representative of three separate experiments. D, TUNEL staining does not occur within 1 hr of stretch. Phase-contrast and fluorescent TUNEL stainimages taken 1 hr after sham (nostretch) or stretch are shown. The data are representative of three separate experiments.

Figure 3.

Stretch plus NMDA-dependent cell death is not caspase mediated. A, Active caspase 3 immunoreactivity in fixed cortical neuronal cultures at 20 hr after the insult using the indicated conditions. Only cultures treated with staurosporine exhibited pronounced active caspase 3 immunofluorescence. Bi, Immunoblot of caspase 3 (both pro and active/cleaved forms) 20 hr after the indicated insult. The pro-caspase 3 form (32 kDa) was detectable under all culture conditions, whereas only cultures treated with stauro sporine displayed the active caspase 3 band (17 kDa). The data are representative of three experiments. Bii, Immunoblot of the time course of caspase 3 expression (both pro and active/cleaved forms) after stretch with or without NMDA. Although pro-caspase 3 (32 kDa) was detectable at all time points, neither stretch nor stretch plus 30 μm NMDA induced active caspase 3 (17 kDa) at any time point. The data are representative of two experiments. C, Stretch-induced vulnerability to NMDA toxicity is not attenuated by the pan caspase inhibitor z-VAD-FMK. Treatment with 200 μm z-vad-FMK for 48 hr attenuated staurosporine-induced death by 47% (t₁₇ = 9.561; p < 0.001). The asterisks indicate the difference from paired control (Bonferroni t test; p < 0.05). The bars show the mean + SE of 9-13 cultures obtained from three separate dissections. D, z-vad-FMK treatment did not reduce DNA laddering (arrowheads) 20 hr after NMDA challenge of stretched cultures (representative of 2 experiments).

Figure 4.

NMDA challenge of stretched neurons does not cause death through caspase-independent apoptotic pathways. A, Lack of nuclear localization of AIF by immunofluorescence. AIF was visualized in fixed cultures at 20 hr after insult. B, Immunoblots of AIF (65 kDa), endo g (35 kDa), and nNOS (160 kDa) in nuclear and cytosolic fractions taken from cells at 6 or 20 hr after insult (stretch ± 30 μm NMDA). All cytoplasmic fractions were positive for immunoreactivity of AIF, endo g, and nNOS. However, nuclear fractions showed only trace immunoreactivity for any of these proteins (representative of 3 separate experiments). C, Inhibiting calpain does not reduce stretch-induced vulnerability to NMDA toxicity. Cultures were prencubated for 30 min with 10 μm calpain inhibitor III and then exposed to the indicated challenge. Calpain inhibitor III remained in the bath until cell death was measured at 20 hr. Calpain inhibition reduced slightly the toxicity of NMDA in all conditions but failed to reduce the vulnerability of the cells to stretch. The asterisks indicate the difference from paired control (Bonferroni t test; p < 0.05). The bars show the mean + SE of 9-12 cultures obtained from three separate experiments.

Figure 6.

Sublethal stretch injury causes extensive ROS production. Cultures were preincubated with DHR (5 μm) for 30 min before insult. DHR fluorescence was measured at 5 min intervals. Statistical comparisons were made at 60 min. A, Applying either 30 μm or 1 mm NMDA to unstretched cultures caused significant ROS production by 60 min compared with controls (30 μm NMDA-t₃₂ = 5.82, p < 0.0001; 1 mm NMDA-t35 = 7.57, p < 0.0001). Stretch alone also caused extensive ROS production (t₃₄ = 15.13; p < 0.0001), similar in magnitude to that of 1 mm NMDA (t₃₁ = 0.817; p = 0.420). Applying 30 μm NMDA to stretched cultures further increased DHR fluorescence compared with stretch alone (t₂₈ = 7.65; p < 0.0001). Symbols represent the mean ± SE of 14-20 cultures obtained from three separate dissections. The error bars are shown where they exceed symbol size. B, Representative DHR fluorescence images at 15 min after the indicated insult. Treating cultures with the mitochondrial uncoupler FCCP did not increase rhodamine-123 fluorescence at 15 min.

Figure 7.

Elimination of stretch-induced vulnerability NMDA toxicity by pretreatment with an ROS scavenger, MnTBAP. Cultures were preincubated for 30 min with 200 μm MnTBAP, which remained in the bath thereafter. Cell death was measured at 20 hr. A, Effects of MnTBAP on cell death under the indicated conditions. The bars represent the mean + SE of 6-12 cultures obtained from three separate dissections. The asterisk indicates the difference from unstretched cultures in same group (t₂₁ = 5.63; p < 0.001). B, Representative phase-contrast and PI fluorescence images of unstretched and stretched cultures 20 hr after the indicated insult. C, Effect of Mn TBAP on ROS levels in unstretched (top) and stretched (bottom) cultures. Pretreatment with MnTBAP reduced the 60 min DHR fluorescence in stretched and unstretched cultures under all insult conditions (Bonferroni t test; p < 0.05). Each symbol represents the mean ± SE of 9-19 cultures obtained from three separate dissections.

Figure 8.

NO production is key instretch-mediated vulner ability to NMDA toxicity. Cell death in A and B was measured at 20 hr. A, Effects of l-NAME, a NOS inhibitor, on cell death under the indicated conditions. Cultures were preincubated with 100 μm l-NAME for 30 min. The asterisk inicates the difference from paired control (t₁₈ = 4.141; p < 0.001). NS, Not different from paired control (t₂₅ = 1.14; p = 0.313). The bars represent the mean + SE of 6-19 cultures obtained from three separate dissections. B, Attenuation of SNP (300 μm; an NO donor) toxicity by the ROS scavenger MnTBAP. The solutions contained MK-801 (10 μm), CNQX (10 μm), and nimodipine (2 μm) to block Ca influx through these pathways. SNP was applied for 1.5 hr. Previous stretch enhanced the vulnerability of neurons to SNP (^*t₁₆ = 5.583; p < 0.001), and this was abolished with a 30 min pretreatment with 200 μm MnTBAP. The bars represent the mean + SE of 6-12 cultures obtained from three separate dissections. C, Effect of NMDA treatment on nitrotyrosine staining at the indicated time and conditions. D, Quantification of nitrotyrosine staining intensity at the indicated times. Background-subtracted fluorescence intensity measurements were taken from 5-15 randomly chosen fields from each culture using identical excitation wavelengths, microscope, and camera settings. The bars indicate the mean + SE of two cultures from each of two separate experiments. The asterisks indicate the difference from unstretched controls at the same time point (Bonferroni t test; p < 0.05).

Figure 9.

Pretreatment with MnTBAP or l-NAME results reduces TUNEL staining and DNA laddering in sublethally stretched cultures challenged with NMDA. Representative images (A) and quantification (B) of TUNEL staining using the DAB method at 20 hr after the indicated insult. The cultures were princubated for 30 min with either 200 μm MnTBAP or 100 μm l-NAME, as indicated. Staurosporine was applied for 48 hr. The asterisks indicate the difference from unstretched control (Bonferroni t test; p < 0.05). Each plotted data represents TUNEL-positive cells normalized to total cell number. The bars represent the mean + SE of two to four randomly selected fields in each of three cultures from each of three experiments. C, Representative DNA gel of the effect of pretreating stretched cultures with either MnTBAP or l-NAME on DNA laddering. The data are representative of three separate experiments.

Figure 11.

Pretreatment and post-treatment with Tat peptides and fusion proteins reduces the stretch-induced increased vulnerability to NMDA toxicity. A, Effect of pretreatment with Tat peptides on survival 20 hr after the indicated insults. Cultures were preincubated with 50 nm Tat peptides for 30 min. The peptides remained in the bath thereafter. Inset, Experimental time course. The asterisks indicate the different from paired control (Bonferroni t test; p < 0.05). N.S., Not significantly different. The bars are the mean + SE of 6-20 cultures obtained from four different experiments. B, Representative phase-contrast and PI fluorescence images of unstretched (left) and stretched (right) cultures 20 hr after challenge with 30 μm NMDA. Pretreatment with Tat-NR2B9c, but not with Tat-NR2B-AA, resulted in decreased PI fluorescence. C, D, Effect of post-treatment with Tat peptides (C) or fusion proteins (D) on cell survival 20 hr after the indicated insult. The peptides or fusion proteins were added 1 hr after insult onset (after termination of the NMDA challenge). Post-treatment with 50 nm Tat-NR2B9c or pTat-PDZ1-2 reduced the vulnerability of neurons to NMDA after stretch. The asterisks indicate the differences from paired controls (Bonferroni t test; p < 0.05). The bars are the mean + SE of 7-22 cultures obtained from four different experiments. Inset, Experimental time course.

Figure 12.

Proposed mechanism of cell death in sublethally stretched neurons exposed to NMDA. Stretch results in increased superoxide production at a level that is still tolerated by the cells. However, subsequent NMDAR activation causes NO production, which permits the formation of peroxynitrite. This, in turn, causes DNA fragmentation by a process independent of classical caspase-dependent apoptosis, caspase-independent apoptosis (AIF, endo g), or caplains.