The role of Nrf2-regulated heme oxygenase 1 on distinct organ responses to early and late fluid resuscitation after hemorrhagic shock
Abstract
Introduction
The therapeutic goals for hemorrhagic shock are to stop bleeding and to restore intravascular volume. Current treatments of HS involve rapid and aggressive fluid resuscitation to restore normal blood pressure and tissue perfusion. However, in recent studies, it had reported that aggressive fluid resuscitation may have adverse effects. Animal studies have revealed that mortality is increased due to attempting to achieve normal blood pressure by aggressive resuscitation during uncontrolled hemorrhagic shock. On the other hand, an increased risk for death was found from aggressive
resuscitation in animals with less severe hemorrhage. This finding suggests that excessive fluid resuscitation can be lethal when severe hemorrhage is not present. Except for the effects of resuscitative volume, different resuscitation fluids also affect the prognosis of HS. Previous study demonstrated that, in prolonged uncontrolled HS, a titrated
intravenous infusion of hypertonic /hyperoncotic solution can maintain controlled hypotension with only one-tenth of the volume of lactate ringer required, without increasing blood loss and edema. However, the causality of different outcomes after resuscitation with different fluid and volume is still unknown. Therefore, our previous studies indicated that resuscitation with lactate ringer and shed blood but not HES (hydroxyethyl starch) solution elevated oxidative stress. These evidences demonstrated that the different prognosis after resuscitation with different fluid and volume might correlate to the change of redox homeostasis.
Although oxidative stress and inflammatory responses has been reported as the causality of HS-induced multiple organs failure, the manifestation of endogenous antioxidant system in response to HS and resuscitation is still unclear. Nuclear factor-erythroid 2 -related factor 2 (Nrf2) is a transcriptional factor binding to the antioxidant response elements (AREs) to regulate the transcription of antioxidant genes, including heme oxygenase 1 (HO-1), and γ-glutamycsteine synthetase (γ-GCS, a key enzyme of GSH biosynthesis). Under basal conditions, Nrf2 is kept inactive through its interaction with the cytosolic protein Kelch-like ECH-associated protein 1 (Keap1). Keap1 is an actin-binding protein that sequesters Nrf2 in the cytoplasm and also
promotes the proteasomal degradation of Nrf2. The negative regulation of Nrf2 activity by Keap1 keeps the basal transcription of ARE-containing genes relatively low. In the presence of oxidants, electrophiles and certain kinases, the Nrf2–Keap1 interaction is disrupted and Nrf2 is released and translocated into the nucleus where it can activate transcription of ARE-containing genes. Therefore, the Keap1-Nrf2 pathway allows for the rapid induction of an array of antioxidant and phase II detoxification genes to combat the initial insult.
GSH depletion), H2O2 exposure, metabolic inhibition by rotenone, and Ca2+ overload.
Materials and methods
Animals:
Studies were performed on 8-week-old male Sprague-Dawley rats (250-350 g). They were maintained on standard laboratory chow and water ad libitum in the animal facility of the Medical College, National Cheng Kung University, Tainan, Taiwan. Rats were raised and cared for according to the guidelines set up by the National Science Council of the Republic of China. The rats were anesthetized with urethane (Sigma). Polyethylene catheters (PE tubing, outer diameter 0.965 mm, inner diameter 0.58 mm, Becton Dickinson and company) filled with 40 IU/mL heparin solution were introduced into the right common carotid artery for blood withdrawal, into the left formal vein for fluid resuscitation, and left formal artery for the measurement of blood pressure. The blood pressure was recorded using a polygraph system (Gould Polygraph 3200). The body temperature of the rats was maintained at 37 ± 0.5 °C by an isothermal pad during the experiments.
Hemorrhagic shock and resuscitation
We induced acute HS using blood withdrawal (25 mL/kg) from the carotid artery catheter and allowed the mean arterial pressure drop to 30-40 mmHg within 30 mins, called onset of HS (HSOS), and maintained the mean arterial pressure in 30-40 mmHg for another 30 minutes, called HS. Withdrawn blood was stored in a heparin pre-rinsed glass bottle at 37°C for the resuscitation. Resuscitation fluid was shed blood or 25% shed blood and 75% Lactate Ringer (3:1, LR+BL). Fluid resuscitation performed at HSOS or HS respectively called early or late resuscitation. Moreover, our model also induced global cerebral ischemia in right hemisphere due to the right common carotid artery occlusion for blood withdrawal and hemorrhagic hypotension. Rats were divided into seven groups: group I, sham operation (Sham); group II, HS treated and killed at the end of blood withdrawal (HSOS); group III, HS treated and killed at the end of HS (HS); group IV, HS and resuscitated with shed blood at the end of blood withdrawal (HS-RI-B); group V, HS and resuscitated with shed blood at the end of HS (HS-RII-B); group VI, HS and resuscitated with 25% shed blood and 75% Lactate Ringer at the end of blood withdrawal (HS-RI-LB); group VII, HS treated and resuscitated with 25% shed blood and 75% Lactate Ringer at the end of HS (HS-RII-LB). The sham group was sacrificed at 2 hours after operation, and the HSOS and HS-RI groups were sacrificed at 30mins after HS on set, and HS and HS-RII groups were sacrificed at 60mins after HS on set.
Blood gas measurement
Blood gas, including pH value and Beecf were determined in blood sample from each group at pre-bleeding as a baseline, HSOS or HS, and before sacrificed using an i-STAT chip (104 Windsor Center Drive-East Windsor, NJ 08520 USA).
Preparation of Nuclear and Cytosolic Extracts
Cytosolic and nuclear fractions were prepared as previously described (Minc, E. et al 1999). Tissues were lysed in lysis buffer(10 mM HEPES,pH 7.9, 1.5 mM MgCl2 ,
10 mM KCl, 1% NP40, 0.5 mM Dithiohtreitol (DTT), 0.5 mM PMSF, 5 μg/ml aprotinin, 4 μg/ml leupeptin, 1 mM NaVO4 ,1 mM Na4P2O7) by homogenizer. Then,
the homogenate was centrifuged at 6100 rpm for 5 min at 4 °C. The supernatant was corresponding to the cytosolic fraction. The nuclear pellet was resuspended in extraction buffer (20 mM HEPES, pH 7.9, 1.5 mM MgCl2, 420 mM NaCl , 0.2 mM EDTA , 25% glycerol, 0.5 mM Dithiohtreitol (DTT), 0.5 mM PMSF, 5 μg/ml aprotinin, 4 μg/ml leupeptin, 1 mM NaVO4 and 1 mM Na4P2O7) and were incubated
for 30 min at 4 °C in a rotating wheel. Nuclear debris was removed by centrifugation at 10700 rpm for 20 min at 4 °C. The supernatant was corresponding to the nuclear fraction.
Electrophoresis mobility shift assay (EMSA) of Nrf2
An aliquot (5 μg) of nuclear protein was incubated on ice with a binding buffer (65 mM Hepes pH 7.9, 250 mM NaCl, 0.5 mM EDTA, 40% glycerol, 0.05 μg PolydI-dC) in 19 μl total volume. After 15 min incubation, 1 μl (30000 cpm) of [γ32P] ATP end-labeled double-stranded oligonucleotide containing an ARE consensus sequence from rat g-GCS promoter (5’ AGC TTG CAC AAA GCG CTG AG T CAC GGG GAG GCG GTG CGC GCC CG 3’) was added to the reaction, and followed by additional 30 min incubation at room temperature. For gel supershift analysis, nuclear proteins (5 μg) were preincubated with 9μl of the anti-human Nrf2 antibody (Santa Cruz) and processed for gel shift assay as described above. The mixture was electrophoresed on a 4% polyacrylamide gel with 1x TGE buffer for 3 h at 150 V and followed by autoradiography.
Western blot and densitometric analysis
Immunohistochemistry and hematoxylin and eosin (H&E) staining
In all groups studied, after HS/R treated, rats were perfused transcardially with 400 ml of 0.1 M (pH 7.4) phosphate buffered saline (PBS) and fixed by 4% paraformaldehyde (pH 7.4). The brains were removed and fixed overnight in the same fixative at 4°C and then submerged in 10%, 20% and 30% sucrose solution for series dehydration in 2 weeks. After dehydration, the brains were serially sectioned at 10μm using a cryostat. Accumulated evidences suggest that NeuN staining may be helpful for quantitative evaluation of neuronal cell loss after brain damage [63-65]. Cryosections (10 μm thick) were mounted on gelatin-coated slides. The sections were incubated with 0.3% H2O2 for 10 min and then with MOM reagent (Vector
Laboratories, Burlingame, Calif.) for 1 h. The sections were then incubated with anti-NeuN antibody (1:200 dilution; Chemicon) overnight at 4°C, followed by peroxidase-conjugated secondary antibody (1:500 dilution) for 1 h at room temperature. The sections were developed with aminoethyl carbazole (AEC substrate kit; Zymed, San Francisco, Calif.) and examined with a light microscope (Nikon eclipse E600). Control sections were incubated with medium instead of anti-neuN.
In order to get a semiquantitative estimation of the number of neurons in the different populations, immunoreacted profiles were counted. The number of neurons in cerebral cortex was counted using light microscopic examination (×200 magnifications) at five representative levels (-4.52 to -4.8 mm from bregma) and estimated using image analysis software (Image Tool, UTHSCSA, San Antonio, TX, USA). The cell count and data analysis were done double blind by two different individuals. H&E staining were performed for the histological examination of liver and lung. Liver and lung segments were fixed in 10 % v/v phosphate-buffered formalin (pH 7.4) for 24 h and then embedded in paraffin. Sections (4 µm) were cut using a microtome, stained with hematoxylin and eosin (H&E) and viewed by light microscopy (Nikon eclipse E600) at ×40, ×200 and ×400 magnification. For each rat, a total of 2 tissue sections were examined, where for each tissue section 5 fields were viewed for the determination of the extent of organ damage. Observers were blinded to the treatment.
Statistical analysis
Results and Discussion
Changes in physiological parameters during HS/R
To identify the effect of hemorrhagic shock and fluid resuscitation (HS/R), we monitored mean arterial blood pressure (MAP), and heart rate (HR), pH value and BE-ecf (base excess of the extracellular fluid) during HS/R. First, we induced hemorrhagic shock by blood withdraw to decrease MAP to 30-40 mmHg and combined with a compensatory increase of HR within 30 minutes, called HS on set (HSOS) (Fig. 1). MAP was further maintained at 30-40 mmHg for another 30 minutes, called HS period (HS) and after HS animals will die without resuscitation (Fig. 1). Because acidosis is the best indicator in early shock of ongoing oxygen imbalance at the tissue level, the HS was also confirmed by a significant decrease of pH and BE-ecf (indicators of acidemia) at time points of HSOS and HS (Fig. 1 C and D, Fig. 2 C and D). Fluid resuscitation at 5 (early) or 20 minutes (late) after HSOS can reverse HS-induced hypotension but not tachycardia (Fig. 1A and B, Fig. 2A and B). We found that early or late fluid resuscitation with 3:1 of L-isomer lactate Ringer’ solution and shed blood (LB) was as potent as shed blood (B) resuscitation to recover HS or HSOS-induced hypotension (Fig. 1B and Fig. 2B). However, early or late fluid resuscitation reversed HS-decreased pH and BE-ecf value (Fig 1C and D, Fig. 2C and D). The results indicated that our HS model indeed worked and fluid resuscitation could reverse HS-induced hypotension and acidemia.
The effects of fluid resuscitation on neuronal loss of HS-induced cerebral ischemia Because common carotid artery occlusion combined with severe hypotension had been used to induce global cerebral ischemia, in the present study, we found severe neuronal loss in ipsilateral cerebral cortex due to the occlusion of right common carotid artery for blood withdrawal and HS induction (Fig. 3C and F). The HS-induced neuronal loss included cortex, hippocampus and striatum of right cerebral cortex and was time-dependent that loss of cortical neurons was severe in HS than HSOS. Our results showed that late fluid resuscitation with shed blood (B) or 3:1 of L-isomer lactate Ringer’ solution and shed blood (LB) significantly rescued HS-induced neuron loss. However, early fluid resuscitation with B or LB did not prevent HSOS-induced neuronal loss (Fig. 3). The results indicated that late fluid resuscitation was an appropriate treatment to prevent HS-induced the death of neurons.
The protein expression of HO-1 after HS and fluid resuscitation
HO-1 upregulation (Fig. 5). Early or late fluid resuscitation with shed blood (B) or LR+BL (LB) prevented HSOS or HS-induced Nrf2 activation (Fig. 4) and HO-1 upregultaion (Fig. 5).
The effects of HS and fluid resuscitation on liver pathology
In contrast to brain, we found that HS only did not cause liver damage; however, early or late fluid resuscitation with LR + BL induced liver damage as indicated by vacuolization, sinusoidal congestion and necrosis in liver (Fig. 6). Moreover, no matter when fluid was resuscitated, shed blood (BL) did not result in liver damage. The results indicated that liver damage after HS/R was dependent on the type of resuscitation fluid.
The protein expression of HO-1 expression after HS and fluid resuscitation in liver We found that there was no significant change of HO-1 expression in HSOS and HS group compared to sham group, but resuscitation with LR+BL significantly increased the HO-1 protein expression (Fig. 7). The HO-1 upregulation after LR+BL fluid resuscitation positively correlated with liver damage. Herein, we proposed that HO-1 protein expression was a sensitive indicator to liver injury.
The effects of HS and fluid resuscitation on lung pathology
Mild lung damage, as indicated by an increase of white blood cells in alveolar, were observed in lungs of groups with fluid resuscitation but not HS only groups. We also found that early fluid resuscitation with LR+BL caused a severe damage in lung with pulmonary alveolar septal thickening and infiltration of inflammatory cells (Fig. 8). Base on histological examination, we found that fluid resuscitation with LR+BL were harmful to liver and lung (Fig. 7 and 8).
The protein expression of HO-1 after HS and fluid resuscitation in lung
HS or fluid resuscitation with LR+BL did not alter HO-1 expression compared with sham group; however, fluid resuscitation with shed blood (BL) significantly downregulated HO-1 expression (Fig. 9). The distinct responses of HO-1 expression in liver and lung after HS/R need to be further investigated.
Conclusion
Figure legends
Figure 1. Changes of mean arterial pressure, heart rate, value of pH and BEecf during hemorrhagic shock and fluid resuscitation at 5 minutes after HS on set. Mean arterial pressure (A), heart rate (B), pH (C) and BE-ecf (D) were monitored during HS and fluid resuscitation at 5 minutes after HS on set. Blood samples (0.8 mL) were obtained from sham, hemorrhagic shock (HS) groups and resuscitation with shed blood (HS-RI-B), or with 3:1 of lactate Ringer’ solution and blood mixture (HS-RI-LB) at 5 minutes after HS on set (HSOS) groups at beginning of blood withdrawn, HSOS and time to sacrifice indicated in upper panel . Values are mean ± SEM, n=6. *, p<0.05, as compared to sham group, #, p<0.05, as compared to the value of HS group at 60 minutes. Black arrows indicate the beginning of blood withdrawn and red arrows indicate beginning of fluid resuscitation.
Figure 2. Changes of mean arterial pressure, pH and BEecf during hemorrhagic shock and fluid resuscitation at 20 minutes after HS on set. Mean arterial pressure (A), heart rate (B), pH (C) and BE-ecf (D) were monitored during HS and fluid resuscitation at 20 minutes after HS on set. Blood samples (0.8 mL) were obtained from sham, hemorrhagic shock (HS) groups and resuscitation with shed blood (HS-RI-B), or with 3:1 of lactate Ringer’ solution and blood mixture (HS-RI-LB) at 20 minutes after HS on set (HSOS) groups at beginning of blood withdrawn, HSOS, beginning of resuscitation and time to sacrifice indicated in upper panel. Values are mean ± SEM, n=6. *, p<0.05, as compared to sham group, #, p<0.05, as compared to the value of HS group at 60 minutes. Black arrows indicate the beginning of blood withdrawn and red arrows indicate beginning of fluid resuscitation.
Figure 3. The effect of fluid resuscitation on HS-induced neuronal loss in right ischemic cortex. Coronal brain sections (bregma, -4.52 to -4.8) from sham (B), 5 minutes after hemorrhagic shock on set (HSOS, C), hemorrhagic shock (HS, F), resuscitation with shed blood (BL) at 5 minutes after HSOS (HS-RI-B, D) or at 20 minutes after HSOS (HS-RII-B, G), or resuscitation with 3:1 of lactate Ringer’ solution and blood mixture (LR + BL) at 5 minutes after HSOS (HS-RI-LB, E) or at 20 minutes after HSOS (HS-RII-LB, H) were stained with anti-NeuN antibody. HS group were sacrificed before death (20-30 minutes after HSOS) and other
site (L).
Figure 4. The effect of fluid resuscitation at different stages of HS on
HS-activated Nrf2 DNA binding activity in cerebral cortex. Gel supershift assay was performed with nuclear protein extract and anti-Nrf2 antibody or competitor to confirm the specific DNA binding of Nrf2 (A). Nrf2 DNA binding activity was
determined by EMSA and represented in the upper panel of B and C in cerebral cortex after HS with or without resuscitation. The treatment and representative symbols are the same as described as Fig. 3. Quantitative data of Nrf2 DNA binding activity by densitometric analysis was shown in lower panel of B and C. n.s is nonspecific binding. Values are mean ± SEM, n=4-9. *, p<0.05, as compared to the left
contralateral cortex (L) of sham group, @, p<0.05, as compared to the corresponding left contralateral cortex of HS or HSOS-treated group, #, p<0.05, as compared to the right ischemic cortex (R) of HS or HSOS treated-group.
Figure 5. The protein expression of HO-1 in cerebral cortex after HS and fluid resuscitation. The HO-1 protein expression in cerebral cortex was determined by western blot shown in upper panel and quantitative data by densitometric analysis was shown in lower panel of A and B. The treatment and representative symbols are the same as described in Fig. 3. Values are mean ± SEM, n=4-9. *, p<0.05, as compared to the left contralateral cortex (L) of sham group, @, p<0.05, as compared to the corresponding left contralateral cortex of HS or HSOS-treated group, #, p<0.05, as compared to the right ischemic cortex (R) of HS or HSOS treated-group.
Figure 6. The effect of HS and resuscitation on liver damage. The histological changes of livers were examined by haematoxylin and eosin stain. Higher microphotographs of A-G were shown in H-N. The ischemic injuries were observed in HS-RILB (E, L) and HS-RIILB (G, N) groups but not in other groups and pyknotic nucleus condense (arrows) was observed J and N.
Figure 7. The protein expression of HO-1 in liver after HS and fluid resuscitation. The HO-1 protein expression in liver was determined by western blot and shown in upper panel and quantitative data by densitometric analysis was shown in lower panel of A and B. The treatment and representative symbols are the same as described in Fig. 3. Values are mean ± SEM, n=4-9. *, p<0.05, as compared of sham group, #, p<0.05, as compared to the corresponding HS or HSOS treated-group.
changes of livers were examined by haematoxylin and eosin stain. Higher microphotographs of A-G were shown in H-N. The increases of white blood cells were observed in every resuscitation group and thickened septa were only observed in HS-RILB groups (L, J).
-15 0 15 30 45 60 75 90 105 120 150 200 250 300 350 400 450 500 Sham HSOS HS-RB 100% HS-RLB Time(min) He a rt r a te ( b p m ) -15 0 15 30 45 60 75 90 105 120 0 20 40 60 80 100 120
*
*
*
*
Time(min) MAP ( mm H g ) 0 30 60 90 6.9 7.0 7.1 7.2 7.3 7.4 7.5 7.6 # # HSOS HS-RB 100% HS-RLB time (min) pH 0 30 60 90 -30 -20 -10 0 10 # # time (min) B E ecf (m m o l/ L ) 60min 5min 30-40mmHg 30min Blood withdraw 30min 25min 0 30 60 90 HS on set Resuscitation I Sacrificed for HS-treated group Sacrificed for HS-R-treated group Bleeding timeStable time HS time Post-resuscitation time
35
Figure 1.
A
B
0 30 60 90 -30 -20 -10 0 10 # # time (min) BE e c f (m m o l/ L ) 0 30 60 90 6.9 7.0 7.1 7.2 7.3 7.4 7.5 # HS HS-RII-B HS-RII-LB time (min) pH # -15 0 15 30 45 60 75 90 105 120 150 200 250 300 350 400 450 500 Sham HS HS-RIIB 100% HS-RIILB Time(min) He a rt r a te ( b p m ) -15 0 15 30 45 60 75 90 105 120 0 20 40 60 80 100 120 Time(min) MA P ( m mH g ) 60min 30-40mmHg 30min Blood withdraw 30-40min 20-30min 0 30 50-60 90 HS on set Resuscitation II Bleeding time
Stable time HS time Post-resuscitation time
（A） L R （C）5min-HSOS （D）HS-RI-B （E）HS-RI-LB （F）HS （G）HS-RII-B （H）HS-RII-LB （B）Sham L R L R L R L R L R L R L R 50μm 0 25 50 75 100 sham HSOS BL HSOS+resuscitation I L R L R L R L R L R LR+BL
