Part 1 Cistanches Alleviates Sevoflurane-induced Cognitive Dysfunction By Regulating PPAR-γ-dependent Antioxidant And Anti-inflammatory in Rats

More information please contact : Joanna.jia@wecistanche.com

Cistanche deserticola has many effects, click here to know more

Abstract

This study aimed to investigate the protective effects and underlying mechanisms of cistanche on the sevoflurane-induced aged cognitive dysfunction rat model. Aged (24 months) male SD rats were randomly assigned to four groups: control group, sevoflurane group, control + cistanche, and sevoflurane + cistanche group. Subsequently, inflammatory cytokine levels were measured by ELISA, and the cognitive dysfunction of rats was evaluated by water maze test, open-field test, and the fear conditioning test. Three days following anesthesia, the rats were killed and the hippocampus was harvested for the analysis of relative biomolecules. The oxidative stress level was indicated as nitrite and MDA concentration, along with the SOD and CAT activity. Finally, a PPAR-γ antagonist was used to explore the mechanism of cistanche in vivo. The results showed that after inhaling the sevoflurane, 24- but not 3-month-old male SD rats developed obvious cognitive impairments in the behavior test 3 days after anesthesia. Intraperitoneal injection of cistanche at the dose of 50 mg/kg for 3 consecutive days before anesthesia alleviated the sevoflurane-induced elevation of neuroinflammation levels and significantly attenuated the hippocampus-dependent memory impairments in 24-month-old rats. Cistanche also reduced oxidative stress by decreasing nitrite and MDA while increasing the SOD and CAT activity. Moreover, such treatment also inhibited the activation of microglia. In addition, we demonstrated that PPAR-γ inhibition conversely alleviated the cistanche-induced protective effect. Taken together, we demonstrated that cistanche can exert antioxidant, anti-inflammatory, anti-apoptosis, and anti-activation of microglia effects on the development of sevoflurane-induced cognitive dysfunction by activating PPAR-γ signaling.

KEYWORDS

cistanche, postoperative cognitive dysfunction (POCD), PPAR-γ, sevoflurane

Cistanche dietary supplements boost immunity

1 | INTRODUC TION

Postoperative cognitive dysfunction (POCD) is a devastating complication with long-lasting consequences and is defined as the impairment of memory, attention, concentration, and information handling occurring after surgery and anesthesia. Research has indicated that about 10% of surgical patients and nearly 40% of elderly patients at age 65 and even older will experience POCD.1 Patient with POCD show strikingly higher mortality, reduced quality of life, and elevated dependency on social security when compared to those without POCD.2,3 Although numerous research efforts had been made in recent decades, the pathogenesis of POCD remains largely unknown.

It is well recognized that neuroinflammation and oxidative stress in the brain play a crucial role in the initiation and progress of POCD.4,5 Although surgery-induced inflammation or oxidative stress can cause neuroinflammation and damage in the brain and ultimately contribute to the cognitive disorder, recent reports suggest that anesthetic agents used during the surgery are also capable of triggering cognitive impairment.3,6,7

Sevoflurane is commonly used in clinical anesthesia for patients of all ages due to its low blood-gas partition coefficient and low metabolic breakdown.8 However, several lines of evidence have shown that sevoflurane could lead to neuroinflammation and impair cognitive function. For example, Makaryus reported that sevoflurane has neurotoxic effects on the development of the brain.9 Decreased connectivity between excitatory neurons in the prefrontal cortex has been shown to be correlated with cognitive impairments induced by sevoflurane.10 Homoplastically, sevoflurane facilitates cognitive decline by aggravating microglia-regulated neuroinflammation in a rat model via down-regulating PPAR-γ activity in the hippocampus.1 Once the surgical method is determined, the degree of trauma and inflammatory response cannot be changed. Therefore, reducing cognitive impairment from the perspective of anesthesia may become a feasible approach. The cistanche species (‘Rou Cong Rong’ in Chinese) has been used as a tonic in China for many years. Modern pharmacological studies have demonstrated that cistanches possess broad medicinal functions, especially on the central nervous system (CNS), including anti-apoptosis, anti-oxidation, anti-aging, anti-fatigue, immunomodulatory anti-inflammatory, and neuroprotection.11-13 Deng et al found that the echinacoside, one of the major active compounds of cistanche, exhibited protective effects on TNF-α-induced SH-SY5Y cell apoptosis.14 Phenylethanoid glycosides, another major active compound of cistanche, which provided abundant phenolic hydroxyl, can serve as a hydrogen donor to reductive radicals and thus scavenge them. There are two main flows by which the cistanches scavenge the free radicals, namely directly involving in the removal of free radicals or blocking their production and regulating the anti-oxidant enzymes related to the free radical metabolism in vivo, such as SOD, CAT, and GPX.15 Cistanches administration contributes to slowing aging phenotypes and relieving cognitive decline in flies by reducing oxidative stress.16 For instance, Zhang et al investigated one of the cistanches extractions and confirmed its function of ex- tending lifespan.17 To date, however, no study examined the effects of cistanches on cognitive dysfunction in either animal models or humans under sevoflurane conditions.

Based on the above findings, we suggested that cistanches could attenuate cognitive impairment induced by sevoflurane. To verify this hypothesis, we used 24-month-old SD rats to get a cognitive impairment model by inhaling sevoflurane. Subsequently, we explored the pharmacological function of cistanches extract on anti-inflammatory, anti-oxidation, and neuroprotection. And the mechanism of cistanches on cognitive impairment induced by sevoflurane was also investigated.

2 | MATERIAL S AND METHODS

2.1 | Preparation of Cistanches Herba extract and standardization

The extraction of cistanches was according to previous works.18,19 Briefly, Cistanches Herba extract was extracted with 95% ethanol in water and lyophilized to the powder. The extract at 10 mg/ ml was applied on quantitative chemical analysis of echinacoside by reverse-phase high-performance liquid chromatography (HPLC). The standard compound, echinacoside, was obtained from Haoxuanbio (Xi'an, Shaanxi, China).

2.2 | Rat model of sevoflurane-induced cognitive dysfunction and treatment with cistanche

Male Sprague Dawley (SD) rats aged 3 and 24 months were purchased from the Beijing Vital River Laboratory Animal Technology Co., Ltd. Rats were raised in clean acrylic cages with free access to water and food under a 12-hour light/dark cycle at a constant temperature of 21°C ± 1°C and humidity (45%-65%) for 1 week before the experiment for accommodation. The animal experimental procedures were authorized by the Animal Care Committee of Seventh People's Hospital of Shanghai University of TCM. All behavioral tests were carried out in a temperature (21°C ± 1°C)- and humidity (45%-65%)-controlled room.

In the first set of experiments in vivo, the 3- and 24-month-old male SD rats were used. Sixteen 3-month-old rats weighing 200- 250 g and sixteen 24-month-old rats weighing 240-300 g were randomly assigned to indraft sevoflurane (Jiangsu Hengrui Medicine Co., Ltd., Lianyungang, China) or carrier gas (n = 8 for each group). Animals in the anesthesia groups (SEVO) were subjected to 2.6% the concentration of sevoflurane delivered by a humidified 30% O2 carrier gas for 4 h at a flow rate of 2 L/min using an anesthetic apparatus with a multi-gas monitor (Prisma SP Alpa, Oxon, UK). Rats in the control group were subjected to the carrier gas without sevoflurane for the same time.

for 4 hours (n = 8 for each group). The behavioral tests were conducted on day 1 and day 3 after anesthesia. Then animals were killed immediately after behavioral assessment on day 3, and hippocampus in the brains was used for molecular analyses. In the third set of experiments, all anesthetized rats were randomly divided into four groups (n = 8 for each group): (a) SEVO group; (b) SEVO + GW9662 (an antagonist of PPAR-γ, 30 ng every time) group;

(c) SEVO + cistanche group; and (d) SEVO + cistanche + GW9662 group. The subsequent experiments were similar to the description in the second set of experiments.

2.3 | Behavioural testing

Morris Water Maze Test was performed as described previously with slight modification to evaluate the spatial memory and learning abilities of rats in the present study.8,20 A tank was filled to a depth of 40 cm with water and made opaque by adding nontoxic white tempera paint. Water temperature was maintained at 22 ± 1°C to prevent the mouse from floating. A circular escape platform (11 cm radius) was placed 0.5-1 cm below the water surface in one of four quadrants. At three time points, rats were tested for reference memory after training for 5 days and each training trial section consisted of four trials with an interval of 15 minutes. At the second and third time-point, the platform was removed from the pool, then spatial and working memory measurement was conducted. The time each rat spent looking for the platform was recorded as the escape latency. After trials, rats were put in heated cages before returning to their cage.

The open-field test was carried out on day 1 and day 3 post-anesthesia as described by Fricano et al21 Briefly, rats were tested in a 27.5 cm × 27.5 cm square arena. The middle, 66% of the arena, was entitled as the ‘center’. The time spent in the center was recorded as the parameter to evaluate the anxiolytic behavior.

The fear conditioning test was performed as described previously with slight modification to explore the spatial memory and learning abilities of rats in the present study.22,23 The training was performed on one day after anesthesia. Each rat was allowed to adapt to the chamber for 120 s, followed by exposure to a sound stimulus (20 s, 80 dB) and then an electric foot shock (2 s, 0.75 mA) 25 s after the sound stimulus ended. The exposures were repeated twice, with 60-s intervals between the second and third exposures. The context test was performed on the second and third days after surgery. Each mouse was put in the same chamber used for training, followed by no stimulus and no foot shock for the same interval used to train the animal. The results were recorded, and the percentage of freezing time (not moving) was collected.

2.4 | Tissue collection

After behavioral analysis at day 3 post-anesthesia, the experimental rats were killed; then, the brain tissues were removed immediately and the hippocampus was dissected carefully. For the proteins, RNAs, and cytokines analysis, the hippocampus was frozen in liquid nitrogen immediately and stored at −80°C. For immunohistochemical analysis, tissues were fixed with 4% paraformaldehyde for 72 hours and then transferred to 20% sucrose for another 72 hours.

2.5 | Enzyme-Linked Immunosorbent Assay

All rats were killed after the behavior test on day 3. Hippocampal tissues were dissected on ice immediately and homogenized in saline by ultrasonic. Then, centrifugation was performed at 10 000 × g at 4°C for 15 minutes. Levels of IL-1β, TNF-α, and IL-6 were determined by enzyme-linked immunosorbent assay (ELISA) kits (eBioscience, San Diego, CA, USA) according to the manufacturer's instructions. Detection limits were 8.0 pg/mL for IL-1β, 8.0 pg/mL for TNF-α and 4.0 pg/mL for IL-6. Plasma IL-1β, TNF-α, and IL-6 concentrations at day 1 and day 3 were also measured using an ELISA.

2.6 | Determination of nitrite/nitrate and oxidative damage to lipids and proteins

The nitrite/nitrate (NO2−) concentration in the collected supernatant was measured by the Griess reaction (Sigma, St. Louis, MO, USA), which showed as an indicator of the production of nitric oxide (NO). Briefly, 100 μL of Griess reagent was added to 100 μL of the supernatant. The total protein content of the supernatant was measured with a BCA protein assay reagent kit (KeyGEN, China) after incubation at room temperature for one hour, and the absorbance was measured using a spectrophotometer at 550 nm. Results are nmol of NO − per mg of protein.

The level of malondialdehyde (MDA) in the hippocampus was determined by the MDA assay kit (Jiancheng Bioengineering Institute, Nanjing, China) after collected via homogenate and centrifugation. The oxidative damage to lipids was determined by the thiobarbituric acid reactive substances (TBARS) assay. Briefly, the supernatant samples were precipitated with 10% trichloroacetic acid. Then thiobarbituric acid (0.67%) was added, and the absorbance was tested at 535 nm. The 1,1,3,3-tetra methoxy propane as an external standard and data were expressed as nmol of malondialdehyde equivalents per mg of protein.

2.7 | Antioxidant enzyme activity detection

For the measurement of superoxide dismutase (SOD) activity, 10 μL of supernatants obtained by tissue homogenizing and centrifugation (10 000 × g, 15 minutes) at 4°C was added to 200 μL reaction solution (50 μL tetrazolium plus 19.95 mL assay buffer, including 50 mM Tris- HCl, 0.1 mmol/L hypoxanthines and 0.1 mmol/L diethylene triamine peracetic acid, pH 8.0). Then, 20 μL of xanthine oxide was added to the mixtures to initiate the reaction. After shaking and incubating for 20 minutes, absorbance changes at 450 nm were recorded using a spectrophotometer. The SOD activities were calculated using the formula supplied with the detecting kit (Cayman Chemical Company, Ann Arbor, MI, USA).

The CAT activity in hippocampal tissue supernatant was determined spectrophotometrically by calculating the hydrogen peroxide (H2O2) decomposition rate at 240 nm. Briefly, an aliquot of 100 μL of the supernatant (20 μL) was added to the substrate mixture (1000 μL) containing 0.3 ml of H2O2 in 50 mL of 0.05 mol/L PBS (pH 7.0). Absorbances were recorded at 1 minute after the initiation of the reaction. The results were calculated using external standards and expressed as U per mg of protein.

2.8 | RNA isolation and quantitative PCR

The hippocampal tissues were homogenized to obtained total RNAs using TRIzol (Invitrogen), and then, transcript amplification was implemented using iScript cDNA Synthesis Kit (Bio-Rad). The resulting cDNA was amplified by qPCR using iTaq Universal SYBR Green Supermix (Bio-Rad) with gene-specific primer primers, and the relative expression of each gene was calculated using the 2−ΔΔCt method and normalized to GAPDH control. The primer sequences are presented in Table S1.

2.9 | Western blot

The hippocampal tissues were homogenized in cell lysis buffer (50 mmol/L Tris, pH 7.5, 1 mmol/L EDTA, 150 mmol/L NaCl, 0.5% Triton X-100, 0.5% NP-40, 1 mmol/L phenylmethylsulphonyl fluoride, 5 mmol/L sodium vanadate, 1 mmol/L sodium fluoride, 1 μg/mL of aprotinin, pepstatin and leupeptin). Proteins were separated by SDS- PAGE gels, and then, proteins in the gels were transferred to PVDF membranes (Millipore). The membranes were blocked with 5% milk and then incubated with specific primary antibodies, including anti- β-APP (ab76763, 1:1000 dilution), anti-BDNF (ab108319, 1:1000 dilution), anti-GFAP (ab68428, 1:1000 dilution), anti-Iba-1 (ab178847, 1:1000 dilution), anti-PPAR-γ (ab209350, 1:1000 dilution), anti-NF-κB (ab32360, 1:1000 dilution) and anti-GAPDH (ab181602, 1:1000 dilution). Subsequently, TBST was used to wash the membranes and then incubated it with HRP-conjugated secondary antibodies (Santa Cruz, CA, USA). Finally, the relevant protein was observed by the ECL system (PerkinElmer) in light of the manufacturer's explication. And ImageJ was applied for quantitative analysis of relative protein expression.

2.10 | Fluorescence-activated cell sorting (FACS) analysis

For FACS analysis, the hippocampal cells were collected after isolation immediately. Then, the cells were washed using ice-cold phosphate-buffered saline (PBS) and resuspended in PBS contain- ing 2% FBS. For each assay, 1 × 105 cells in 100 µL PBS-FBS were stained with fluorescently conjugated CD11b (BD PharMingen, catalog #554982, 1:250) and CD45 (BD PharMingen, catalog #554878, 1:250) antibodies on ice for 2 hours at 4°C in the dark. After washing with cold PBS, cells were resuspended in 300 µL PBS-FBS. And flow cytometric analysis was performed on a FACSAria I flow cytometer (BD Biosciences). Results were analyzed using FlowJo (version 7.6.1).

2.11 | Nissl staining

The collected hippocampus was embedded in an OCT compound and then cut into 14-μm coronal sections using a freezing microtome (Leica). After thaw-mounting on charged slides, the sections were washed with 0.01 mol/L PBS three times for a total of 15 minutes and then incubated in cresyl violet solution (0.5%, containing a few drops of glacial acetic acid) for 10-15 minutes at room temperature. After washing with distilled water twice, the sections were dehydrated in ethanol gradually (70%, 85%, 95%, and 100%). Following dehydration, they were placed in xylene and coverslipped using a mounting medium. Finally, the Nissl staining was observed under the micro- scope (Leica).

2.12 | Data analysis

Data were analyzed using the GraphPad Prism 6.0 software. Data in the behavior studies were analyzed at different time points (day 1 or day 3) after sevoflurane anesthesia using a two-way analysis of variance (ANOVA) with repeated measures. Other data were analyzed using one-way ANOVA before the Bonferroni test. All experiments were performed in triplicate, and the results were expressed as the mean ± SEM P < .05 was considered statistically different.

3 | RESULTS

3.1 | Cistanche prevented abnormal neuroinflammation in aged rats

To investigate the difference in cognitive function between adult and aging rats, sevoflurane was used to induce anesthesia on the 3- and 24-month-old rats. Then, the hippocampus-dependent behavior was evaluated on day 1 and day 3 (Figure S1A). As shown in Figure S1B, there was no difference in escape latency for four experimental groups on day 1. Furthermore, the time spent in the center and the percent of freezing were similar to rats that received vehicle and sevoflurane on day 1 (Figure S1C, D). However, the 3- and the 24- month-old rats had significant differences in the escape latency, time spent in the center and the percentage of

freezing after the sevoflurane induction on day 3, suggesting that sevoflurane altered the cognitive function of aged animals significantly (Figure S1B-D). Thus, the aged rats were used for the subsequent experiments.

To investigate the effects of cistanche on neuroinflammation in aged rats undergoing sevoflurane exposure, we evaluated the levels of the pro-inflammatory cytokine. The evidence for neuroinflammation was seen at 24 hours after sevoflurane anesthesia, as demonstrated by a high concentration of pro-inflammatory cytokines IL-1β, IL-6, and TNF-α in the plasmid on day 1 and day 3 (Figure 1A, 1), and the neuroinflammation was indicated by the elevated mRNAs of pro-inflammatory cytokines in the hippocampus up to 3 days (Figure 1C-E). For each of the pro-inflammatory factors, cistanche treatment eliminated the differences in the sevoflurane-induced changes in the proinflammatory mediators. Cistanche significantly attenuated the elevated pro-inflammatory cytokines in the sevoflurane

3.2 | Cistanches Herba extract treatment attenuated sevoflurane-induced oxidative stress in the hippocampus of aged rats

Next, we investigated the effects of cistanche on the oxidative stress induced by sevoflurane. As illustrated in Figure 2A, we found that the cistanche treatment prevented sevoflurane-induced oxidative stress, as suggested by significantly decreased concentration of nitrite/nitrate in the hippocampus. We also observed that the SEVO + cistanche group presented lower levels of MDA equivalents, an index of the oxidative damage to lipids, in the hippocampus at three days after the sevoflurane inhalation, compared to the SEVO

group (Figure 2B). In addition, rats subjected to sevoflurane anesthesia (SEVO group) presented a significant decrease in SOD and CAT activity on day 3 after anesthesia in the hippocampus. However, the hippocampal activity of those enzymes was significantly increased in SEVO + cistanche group, compared to the SEVO group (Figure 2C, D).

Cistanche Dietary Supplement Relieves Antioxidative Stress

3.3| Cistanches Herba extract treatment attenuated microglial activation

Microglial cells are activated during the laboratory or anesthesia-induced inflammation in the hippocampus, which then contributes to the pathological progress. To investigate whether cistanche could prevent the microglia activation, we measured the population of activated microglia cells and its relative markers. Resting microglia expressed CD11b and a low expression of CD45, while activated microglia expressed CD11b and high level of CD45. As presented in Figure 3A, results from flow cytometry analysis confirmed that the percentage of CD45 was significantly lower in cistanche-treated rats than in SEVO-induced rats (0.495 ± 0.051% following cistanche treatment vs 1.127 ± 0.032% in the SEVO group). Subsequently, we measured the glial markers by qRT-PCR. After anesthesia with sevoflurane, the mRNA

3.4 | Cistanches Herba extract treatment attenuated sevoflurane-induced neuronal apoptosis

Neuronal apoptosis is partially driven by activated microglia and oxidative stress, and we had confirmed that cistanche could prevent microglia activation and oxidative stress in vivo. To further investigate the pharmacological effects of cistanche, we tested for neuronal apoptosis with TUNEL staining and cleaved-caspase 3 and Bax/Bcl-2 immunoblot. As presented in Figure 4A, sevoflurane increased cell apoptosis in the hippocampal regions. Cistanche administration reduced TUNEL+ cells following sevoflurane anesthesia. We also illustrated the remarkable high levels of caspase 3 and Bax, while low levels of Bcl-2 in the hippocampus of anesthetized aged rats. Rats that received administration of cistanche before anesthesia revealed a significant reduction of caspase 3 activity (Figure 4B), cleaved caspase 3, and Bax protein (Figure 4C, D), which was consistent with reduced neuronal apoptosis observed in TUNEL staining.

3.5 | Cistanches Herba extract treatment alleviated cognitive dysfunction and impairment of hippocampal tissue following sevoflurane-induced anesthesia

To further investigate the role of cistanche on cognitive function, rats were subjected to 50 mg/kg of cistanche orally for three consecutive days prior to anesthesia. In the cistanche-treated group (group SEVO + cistanche), the escape latency was significantly shortened, the time spent in the center was longer, and the percent of freezing was increased at day 3 after anesthesia when compared to group SEVO (Figure 5A-C). These results indicated that cistanche could at least partially rescue cognition impairments. We further evaluated the extent of neuronal death in the hippocampus of rats in four groups by Nissl staining. Images presented that cells had no obvious abnormality in group control and group control + cistanche. However, the number of survival neurons in the CA1 regions was significantly reduced in the SEVO-induced rats compared with that in the control group on day 3. Treatment with cistanche reversed this effect of SEVO and significantly promotes neuronal survival compared with SEVO-induced alone (Figure 5D).

3.6 | Cistanches Herba extract treatment reduced sevoflurane-induced pro-inflammatory cytokine production in a PPAR-γ-sensitive manner

We next investigated whether the effect of cistanche on cytokine production was PPAR-γ-sensitive. Treatment with GW9662 (group SEVO + GW9662) increased the levels of IL-1β, IL-6, and TNF-α compared to the SEVO group. However, treatment with cistanche diminished SEVO-induced pro-inflammatory cytokine production. Of note,

3.7| Cistanches Herba extract treatment inhibited sevoflurane-induced oxidative stress via PPAR-γ

It had been reported that sevoflurane exaggerates cognitive dysfunction in a chronic intermittent hypoxia-induced rat model through inhibition of PPAR-γ expression in the hippocampus.1 We next wondered whether the effect of cistanche occurred in a PPAR-γ-dependent manner. For this purpose, GW9662, a selective and irreversible antagonist of PPAR-γ, was used for further study. Treatment with GW9662 at 30 ng promoted oxidative stress, as demonstrated by the increased levels of nitrite/nitrate and MDA. However, the anti-oxidative markers SOD and CAT in hippocampus on day 3 were increased compared to SEVO-operated rats. Treatment with cistanche reduced SEVO-induced overproduction of oxidative stress markers and facilitated the SOD and CAT activity (Figure 7A-D). Furthermore, these effects produced by cistanche were reverted by GW9662, as observed by an increase in nitrite/ nitrate (Figure 7A) and MDA (Figure 7B) production. In addition, the

effects of cistanche on SOD and CAT activity were also reverted by GW9662 (Figure 7C, D), indicating the participation of PPAR-γ.

3.8 | Cistanches Herba extract treatment inhibited microglial activation and neurocyte apoptosis and ameliorated the cognitive decline in a PPAR-γ- sensitive manner

The next step was to explore whether the effects of cistanche on microglia activation and neurocyte apoptosis were PPAR-γ-sensitive. Treatment with GW9662 increased the mRNA levels of CD68 (Figure 8A), GFAP (Figure 8B), and Iba-1 (Figure 8C). Notably, these effects produced by GW9662 were effectively reverted by cistanche, as observed by a decrease in these same markers (Figure 8A-C). Moreover, the protein expression of GFAP and Iba-1 produced by GW9662 was also reverted by cistanche, indicating that cistanche inhibits microglia activation, and at least partially depends on PPAR-γ. Additionally, the effect of cistanche on β-APP and BDNF protein expression was inhibited by GW9662 (Figure 8D, E). To investigate whether cistanche could inhibit neurocyte apoptosis in a PPAR-γ-sensitive manner, apoptosis-related proteins were detected. As shown in Figure 8F, G, cistanche at 50 mg/kg reduced the expression of pro-apoptosis protein (cleaved-caspase 3 and Bax). However, these effects were largely rescued by GW9662.

Cistanche extract can alleviate cell apoptosis

Since the oxidative stress and inflammation levels were closely related to cognitive function in post-anesthesia rats, we further determined whether the effect of cistanche on cognitive dysfunction was PPAR-γ-dependent. As shown in Figure 8H-J, compared with rats