Hydrogen sulfide ameliorates isoflurane-induced cognitive impairment in mice: Implication of caspase-3 activation

Purpose: Isoflurane could induce cognitive impairment and activate caspase-3. However, the mechanism of action is unclear and target interventions are unavailable. The present study examined the potential protective function of hydrogen sulfide (H2S) against isoflurane-induced cognitive impairment. Methods: Effects of NaHS (5 mg/kg) on cognitive impairment induced by isoflurane (1.4% for 2 h) were assessed using a fear-conditioning test in a group of 8-month old mice. H4 human neuroglioma cells, which were transfected with upregulated human amyloid precursor protein were treated for 3 or 6 h with 2% isoflurane, in the presence of 100-μM NaHS in the mice. A group of mice treated with normal saline in place of the NaHS in each case served as control. Western blotting, fluorescence assay, and a mitochondrial swelling assay were employed to observe the results of caspase-3 activation, mitochondrial dysfunction, and ROS and ATP levels. Results: NaHS significantly mitigated isoflurane-induced cognitive impairment in mice. In cultured cells, NaHS reduced caspase-3 activation, ROS, mitochondria membrane reduction, mitochondrial permeability transition pore opening, and cellular ATP level. NaHS could ameliorate cognitive impariment induced by isoflurane through inhibiting caspase-3 activation, oxidative stress, and mitochondrial dysfunction. Conclusion: These results indicate that hydrogen sulfide (H2S) has potential protective function against isoflurane-induced cognitive impairment. Further investigation of NaHS as an intervention to attenuate anesthesia-associated neurotoxicity is vital.


INTRODUCTION
Alzheimer's disease (AD) is a prevailing reason for dementia in adults [1]. Symptoms typically start at age of 60 years and might involve memory loss, failure in performing daily activities, and personality change. The evidence so far shows that exposure to general anesthetics could interact with AD-associated pathological mechanisms, and promote the progression of AD [2].
A randomized controlled trial showed that sevoflurane could accelerate the progression of cognitive impairment, which often precedes AD [3]. A case-control study in China established that anesthesia during surgery is associated with the risk of dementia [4]. Another study indicated that people in the middle age without cognitive impairment are more likely to have rapid cognitive decline after surgery and anesthesia [5]. Several other studies, however, failed to establish this association in anesthesia and dementia [6][7][8]. It was previously reported that isoflurane might promote caspase-3 activation [9], increase amyloid beta-protein (Aβ) level in the brain [10][11][12], and impair learning and memory [13]. The mechanism on how isoflurane enhances caspase activation is not completely understood, but may result from increased levels of ROS, mitochondrial dysfunction and decreased ATP level [14,15].
In this study, we hypothesized that sodium hydrosulfide (NaHS) might inhibit caspase-3 activation via ROS-, mitochondria-, and ATPassociated mechanisms in glioma cells transfected with amyloid precursor. Therefore, the purpose of this study was to explore the underlying mechanisms of caspase-3 activation in vitro and examine the potential effect of NaHS on cognitive impairment in isoflurane-induced mice model.

In vivo Experiments
Following ethical approval by the Animal Care and Use Committee of Shanghai Tenth People's Hospital, Tongji University School of Medicine, this study was carried out following the Regulations of Experimental Animal Administration. Wild-type C57BL/6J mice (8 months old) of both sexes (20 females, and 20 male, with 5 males and 5 females in each group) were placed under a reversed 12h/12h light/dark cycle at 20-22C for a period of 1 week before the experiments in Shanghai Laboratory Animal Center, Chinese Academy of Sciences, Shanghai, China. At random, each mouse in the treatment group (10 mice; with 5 males and 5 females) received an intraperitoneal injection of 5 mg/kg NaHS (Cat #: 161527, Sigma-Aldrich, St Louis, MO, US) while each mouse in the control group (10 mice; with 5 males and 5 females) received injection of 0.1 ml saline. The dosage of NaHS used was based on previous studies, which showed that 5 mg/kg for rats would have desirable results [28][29][30][31][32]. Thirty minutes after NaHS or saline dose, the mice were exposed to 1.4% isoflurane in 100% O2 for 2 hr, using a home-made chamber. Isoflurane concentration was verified, and adjusted when needed using a gas analyzer (GE Healthcare, MA, USA). All animals were consistently kept at 37 ± 0.5C during the experiments.
Fear conditioning test was carried out as described previously [15], with slight modifications using a standard chamber from Stoelting (Wood Dale, IL, USA). On the first day after isoflurane, mice were habituated for 3 min before the presentation of the auditory tone (60 s, 3600 Hz, 80 dB). In the final 2 seconds of the tone, mice received gentle foot shock (0.8 mA, 0.5 sec) through the floor grid. After the foot shock, mice were kept in the testing chamber for 30 seconds before sending to the home cage. On the second day, mice first received a context test (without any stimulus). Two hours later, a tone test was carried out. Briefly, mice were placed in a novel context (distinct wall color and odor) for 3 min before the presentation of the tone for 3 min. During the context and tone experiments, the time during which the mice exhibited freezing behavior (failure to make any movement except respiration) was noted and evaluated with an Any-Maze video tracking system with the following settings: first, freezing on threshold for 10 sec and then freezing off threshold for 20 sec and last minimum freezing duration for mere 1 sec). The chamber was cleaned with 75% ethanol after each session.

Western blot assay
Cells were lyzed using a buffer containing 150 mM NaCl, 10 mM Tris-HCl, 0.5% Nonidet P-40, 2 mM EDTA, 1 μg/mL leupeptin, 1 μg/mL aprotinin, and 1 μg/mL pepstatin A [15]. Lysate obtained was then centrifuged at 2500×g for 15 min and the supernatant was probed using a polyclonal rabbit-anti-mouse anti-caspase-3 antibody (Cat #: 9662, 1:1100; Cell Signaling Technology, MA, USA) that recognizes the full-length (35-40 kDa) and 17-20 kDa caspase-3. Anti-rabbit IgG (#7074, 1:5000; Cell Signal Technology) was used as secondary antibody. An enhanced chemiluminescence method (Thermo Fisher, USA) was employed and then analyzed on NIH ImageJ (US National Institutes of Health, USA). The results obtained were normalized with the βactin control. Protein concentration was measured using a bicinchoninic method. All experiments were done six times and the mean was reported. .

ROS measurement
ROS concentration was measured using OxiSelect Intracellular ROS Kit and an OxiSelect In Vitro ROS/RNS Kit (Cell Biolabs, CA, USA) as described previously [15]. The fluorescence value was maintained at 480/530 nm.

Isolation of mitochondria
Active Motif kit (Carlsbad, CA, USA) was used to isolate the mitochondria. Cell homogenate and the supernatant were centrifuged at 800 × g and 10,000 × g, at 4°C for 20 min to obtain the mitochondrial (pellet) and a cytosolic fraction (supernatant), respectively [33].

Evaluation of MMP
Mitochondrial membrane potential (MMP) was estimated through tetraethylbenzimidazolylcarbocyanine iodides (JC-1) fluorescence ratio detection and tetramethylrhodamine ethyl ester and perchlorate (TMRE). JC-1 MMP detection kit (30001, Biotium, CA, USA) were used as described previously [15]. MMP was calculated as the ratio of red fluorescence to green fluorescence by fluorometric plate reader (Thermo Fisher Scientific, MA, USA). For measuring MMP by TMRE, cells were treated and evaluated under a 100 X objective lens fluorescence microscope (BZ 9000, Keyence, Itasca, IL, USA).

ATP measurement
ATP (A22066 Invitrogen, CA, USA) concentration was examined as described previously [15]. Briefly, to calculate the ATP, fluorescence was measured and compared with a curve obtained from samples of known ATP.

Data analysis
Data are presented as mean values ±SD and analysed using ANOVA and post-hoc Bonferroni as appropriate. At 95% confidence interval, p<0.05 was regarded to be statistically significant.

NaHS ameliorates cognitive impairment induced by isoflurane
FC tests showed that mice in isoflurane group had decreased freezing time to context ( Figure  1A) (F = 12.027, P < 0.0001) and to tone ( Figure  1B) (F =6.482, P = 0.004) compared to control group. Decreased freezing time suggests that the isoflurane induced the cognitive deficit. Furthermore, in the isoflurane NaHS group, the decreased freezing time was alleviated compared to isoflurane group ( Figure 1A-B).

NaHS inhibits Caspase-3 activitycaused by isoflurane
WB showed increased caspase-3 fragments, but not full-length caspase-3, upon isoflurane exposure (p=0.034) (Figure 2). NaHS attenuated the increase of caspase-3 fragments induced by isoflurane (p = 0.046) but did not affect caspase-. Figure 1: NaHS ameliorates isoflurane-induced cognitive impairment. A: Results of the FCT context test in mice treated with control conditions plus saline, isoflurane (1.4%) plus saline, control conditions plus NaHS (5 mg/kg) and isoflurane (1.4%) plus NaHS (5 mg/kg) for 2 h. B: Results of the FCT tone test in mice treated with control conditions plus saline, isoflurane (1.4%) plus saline, control conditions plus NaHS (5 mg/kg) and isoflurane (1.4%) plus NaHS (5 mg/kg). The two-way ANOVA and a post hoc Bonferroni test were used to analyze the results 3 fragment in cells not exposed to isoflurane. Isoflurane and NaHS had significant interaction (F = 5.847, p = 0.0076).

NaHS attenuates ROS increase induced by isoflurane
Isoflurane and NaHS had a significant effect on ROS level (F = 34.12, p < 0.001) (Figure 3). NaHS attenuated the increase in isofluraneinduced ROS level (p < 0.001) and decreased ROS in cells not exposed to isoflurane (p < 0.001).

NaHS attenuates isoflurane-induced mPTPs opening
Ca 2+ induced significant mitochondria swelling in isoflurane-exposed cells (p<0.01) NaHS attenuated Ca 2+ -induced mitochondria swelling and permeability transition (P<0.01 for isoflurane) but did not affect mitochondria swelling in cells not exposed to isoflurane ( Figure  4).

DISCUSSION
The FCT results demonstrate that NaHS could mitigate isoflurane-induced cognitive impairment. An experiment in cultured cells showed that NaHS mitigated caspase-3 activation induced by isoflurane, ROS accumulation, ATP reduction, and mitochondrial dysfunction. This may be the first study showing NaHS could inhibit isofluranemediated cytotoxicity whose underlying mechanism of reduced cytotoxicity could be attributed to suppression of isoflurane-induced ROS accumulation, ATP reduction, and mitochondrial malfunction.
Oxidative stress often results in different neurodegenerative diseases. It was reported that H2S provides protection to primary neurons and immortalized mouse hippocampal cells from  oxidative glutamate toxicity [19,22]. Additionally, Lu et al. discovered that H2S enhances glutamate uptake and increases glutathione B production to prevent excessive glutamate accumulation in synaptic clefts, thus protecting astrocytes from oxidative toxicity [23]. Similarly, our findings indicate that NaHS could attenuate ROS accumulation in cells induced by isoflurane, suggesting that NaHS could mitigate the oxidative toxicity of isoflurane.
The mPTP opening decreases the potential of the membrane and releases apoptogenic proteins to amplify the apoptotic mechanism [35][36][37][38]. Oxidative and other cellular stresses cause mPTP opening [39]. In contrast, inhibition of mPTP opening protects neurons from apoptosis [34]. Hu et al. have shown that H2S inhibits apoptosis by rotenone in SH-SY5Ycells by protecting mitochondria [21]. Xie et al. proposed that mPTP opening and MMP reduction contribute to caspase-3 activation by isoflurane, which can be attenuated by a hydrogen-rich saline [15,39]. Our findings are consistent with these findings as they show that NaHS inhibited isoflurane-induced mPTP opening, and thus attenuated caspase-3 promotion by isoflurane. These results supported the notion that caspase-3 promoted by isoflurane involved mitochondrial dysfunction and ROS mechanisms.
Our FCT experiments in mice demonstrated that NaHS could attenuate cognitive impairment caused by isoflurane exposure. This finding encourages future investigation of NaHS as a treatment for neurobehavioral decline induced by anesthesia. Various behavior studies have shown that H2S could ameliorate cognitive and memory impairments by attenuating ROS production and the activation of apoptotic pathways. Previous studies using Morris maze test showed that H2S (i.e., NaHS) or Spropargylcysteine (SPRC), could ameliorate lipopolysaccharide-induced cognitive impairment [28,40]. Chu et al. showed that NaHS could alleviate memory impairment in mice due to surgery [30]. A study using Aβ rat models of AD showed that NaHS could attenuate spatial memory impairment [29]. He et al. demonstrated that NaHS improved memory acquisition and spatial learning in APP/PS1 transgenic mice [25]. Altogether, these results indicated that exogenous H2S could attenuate cognitive dysfunction due to a variety of etiological factors.
First important limitation of this study was that we did not examine the dose-response relationship as Yin et al. showed that NaHS protected from I/R injury by regulating oxidative stress in a dosedependent manner (e.g., 0.2 and 0.4 μmol/kg) [41]. It is possible that, at different doses, NaHS may have distinct or even opposing effect on caspase-3 activity induced by isoflurane, ROS, mitochondria, and ATP. Second, the behavioral effects of NaHS were examined using only FCT. Third, we did not check the potential functions of different H2S donors (e.g. SPRC). Effects of H2Sreleasing agents with sustained and controlled release that more faithfully mimic physiological conditions are essential. Despite these limitations, the present study showed that NaHS could attenuate cognitive impairments induced by isoflurane exposure.

CONCLUSION
This study indicated that NaHS could ameliorate isoflurane-induced cognitive impairments, likely through attenuating caspase-3 activation induced by isoflurane, mPTP opening, ROS increase, and MMP reduction. Future studies using other H2S donors (e.g., SPRC) and other anesthetics (e.g., sevoflurane) are warranted.