Oxygen-sensitive outcomes and gene expression in acute ischemic stroke

Abstract

Acute ischemic stroke (AIS) results in focal deprivation of blood-borne factors, one of them being oxygen. The purpose of this study was two-fold: (1) to identify therapeutic conditions for supplemental oxygen in AIS and (2) to use transcriptome-wide screening toward uncovering oxygen-sensitive mechanisms. Transient MCAO in rodents was used to delineate the therapeutic potential of normobaric (NBO, 100% O(2), 1ATA) and hyperbaric oxygen (HBO, 100% O(2), 2ATA) during ischemia (iNBO, iHBO) and after reperfusion (rNBO, rHBO). Stroke lesion was quantified using 4.7 T MRI at 48 h. Supplemental oxygen during AIS significantly attenuated percent stroke hemisphere lesion volume as compared with that in room air (RA) controls, whereas identical treatment immediately after reperfusion exacerbated lesion volume (RA=22.4+/-1.8, iNBO=9.9+/-3.6, iHBO=6.6+/-4.8, rNBO=29.8+/-3.6, rHBO=35.4+/-7.6). iNBO and iHBO corrected penumbra tissue pO(2) during AIS as measured by EPR oxymetry. Unbiased query of oxygen-sensitive transcriptome in stroke-affected tissue identified 5,769 differentially expressed genes. Candidate genes were verified by real-time PCR using neurons laser-captured from the stroke-affected somatosensory cortex. Directed microarray analysis showed that supplemental oxygen limited leukocyte accumulation to the infarct site by attenuation of stroke-inducible proinflammatory chemokine response. The findings provide key information relevant to understanding oxygen-dependent molecular mechanisms in the AIS-affected brain.

Figure 1

Supplemental oxygen therapy reduces during, exacerbates at reperfusion, the AIS infarct volume. (A) Oxygen therapy treatment groups. (B) Mean local cerebral blood flow across treatment groups from the stroke-affected MCA area (−0.1 mm bregma, +2.0 mm lateral) at the onset of ischemia (white bar) and upon reperfusion (black bar) relative to baseline. Successful MCAO was validated by a >70% reduction in the MCA-area blood flow. (C) T2-weighted MRI taken 48 h after MCAO under RA, iNBO, iHBO, rNBO, and rHBO conditions. Color look up table applied—shift from blue to red denotes edema and stroke-induced infarct. One representative image from each group. (D) Percent hemispherical infarct volume corrected for edema calculated from T2-weighted MR images (n⩾5). (E) Gait disturbance testing performed 48 h after MCA reperfusion to assess sensorimotor function. ^*P<0.05 versus RA; ^†P<0.05 versus iNBO, iHBO.

Figure 2

Oxidative stress and neurodegeneration after supplemental oxygen therapy in AIS. Representative × 20 FOV images of lipid peroxidation (A, 4HNE), protein oxidation (B, NT), nucleic acid oxidation (C, 8OHdG), and neuron-specific degeneration marker FJ (D) in the contralateral and stroke-affected (RA, iNBO, iHBO, rNBO, rHBO) S1 cortex 48 h after MCAO. Bar=100 μm. (E) A representative coronal section showing the contralateral (c) and stroke (s) hemisphere S1 cortex after 4HNE staining from which × 20 FOV images were captured. (F–I) Quantification of expression (% area) was determined from the S1 cortex (n⩾5) for 4HNE, NT, 8OHdG, and FJ. ^*P<0.05 versus c; ^†P<0.05 versus RA. Bar=100 μm.

Figure 3

Supplemental oxygen therapy corrects penumbral tissue pO₂ during MCAO. (A) A LiNc-BuO probe was stereotactically implanted into the S1 cortex using the following coordinates: −0.1 mm bregma, +2.0 mm lateral, −1.0 mm dorsal. Probe placement was assessed using MRI (arrow). (B) One week after probe delivery, EPR spectra were acquired and averaged from 60-s scans (n=10) at baseline (before MCAO) and during MCAO in animals subjected to RA, iNBO, and iHBO conditions (n⩾3). (C) The peak-to-peak width of the EPR spectra was used to calculate pO₂ using a standard calibration curve. ^*P<0.05, baseline versus RA; ^†P<0.05, RA versus iNBO.

Figure 4

Microarray analysis of the oxygen-sensitive transcriptome in AIS. (A) A flow chart depicting the microarray analysis process. After normalization and Benjamini–Hochberg false discovery correction, 5,769 differentially expressed probesets were identified assuming a twofold or greater increase/decrease in expression and a significance value of P<0.05. (B) Heat-map visualization of differentially expressed probes—2,488 probes with lower expression and 3,281 probes with higher expression in iHBO-treated stroke tissue as compared with that in RA-treated tissue. (C) A volcano plot depicting the rodent transcriptome, with the oxygen-sensitive subset highlighted in yellow.

Figure 5

Laser microdissection pressure catapulting and real-time quantitative–PCR validation of microarray targets. (A and B) Neurons of S1 cortex were rapidly stained for anti-NeuN and captured (arrows) by laser microdissection pressure catapulting. (C) To verify the specificity of the captured elements, gene expression of a neuronal (NF-L) and a glial marker (GFAP) was checked by real-time PCR. (D–G) Expression of oxygen-sensitive candidate genes was validated by real-time PCR in the stroke-affected (ischemic) and contralateral (control) neurons from RA and iHBO animals at 12- and 24-h time points. (D) MCP1, (E) interleukin-6, (F) neural cell adhesion molecule-L1, and (G) glutamate oxaloacetate transaminase. <ND>, not detectable. ^*P<0.05, RA stroke versus iHBO stroke at a given time point; ^†P<0.05, control versus stroke within the oxygen therapy group at a given time point; ^#P<0.05, same group, the 12- versus the 48-h time point.

Figure 6

Supplemental oxygen during AIS attenuates proinflammatory leukocyte accumulation in stroke-affected tissue. Representative × 20 FOVs from the stroke-affected S1 cortex show increased intensity and number of MPO⁺ leukocytes (brown) in RA-treated animals (A) as compared with that in iHBO-treated animals (B). The black arrow indicates leukocyte at × 20 selected for the enlarged view in the inset. MPO⁺ leukocytes were not detected in contralateral control tissue (not shown). (C) Quantification of MPO⁺ leukocytes from representative control and stroke FOVs of RA- and iHBO-treated animals (n⩾5). (D) The relative gene expression of NADPH oxidase subunit p47phox, a key mediator of leukocyte-mediated respiratory burst, in the stroke-affected cortex laser-captured from iHBO brain tissue as compared with that in the RA control. ^*P<0.05, iHBO versus RA; ^†P<0.05, RA stroke versus RA control. Bar=100 μm.