The Seipin gene was originally found to be responsible for type 2 congenital lipodystrophy and involved in lipid droplet formation. Seipin is highly expressed in the central nervous system as well. Seipin mutations have been identified in motor neuron diseases such as Silver syndrome and spastic paraplegia. In this study, we generated neuron-specific seipin knockout mice (seipin-nKO) to investigate the influence of seipin deficiency on locomotion and affective behaviors. In comparison with control mice, 8-week-old male seipin-nKO mice, but not female mice, displayed anxiety- and depression-like behaviors as assessed by open-field, elevated plus-maze, forced swim and tail suspension tests. However, neither male nor female seipin-nKO mice showed locomotion deficits in swimming tank and rotarod tests. Interestingly, the mRNA and protein levels of peroxisome proliferator-activated receptor gamma (PPARγ) in the hippocampus and cortex were lower in male seipin-nKO mice, but not female mice, than controls. In seipin-nKO mice, plasma levels of sex hormones including 17β-estradiol (E2) in females and testosterone in males as well as corticosterone were not altered compared with controls. The treatment of male seipin-nKO mice with E2 ameliorated the anxiety- and depression-like behaviors and remarkably increased PPARγ levels. The PPARγ agonist rosiglitazone alleviated affective disorders in male seipin-nKO mice. Notably, anxiety- and depression-like behaviors appeared in female seipin-nKO mice after ovariectomy, which was associated with low PPARγ expression. Collectively, these results indicate that neuronal seipin deficiency causing reduced PPARγ levels leads to affective disorders in male mice that are rescued by E2-increased PPARγ expression.
Congenital generalized lipodystrophy (CGL) is an autosomal recessive disorder characterized by a near-total loss of adipose tissue, severe insulin resistance, hypertriglyceridemia and fatty liver (1). Genome-wide linkage analysis has identified two loci for CGL: CGL1 by mutation in the 1-acylglycerol-3-phosphate-O-acyl transferase 2 (AGPAT2) gene and CGL2 by mutation in the Berardinelli-Seip congenital lipodystrophy 2 (BSCL2) gene that encodes seipin (2,3). Three species of seipin mRNA at ∼1.6, 1.8 and 2.2 kb are found in humans (4). The 1.8 kb mRNA is exclusively expressed in the brain and testis, whereas the other two mRNA transcripts are ubiquitously expressed (5). Considering its broad tissue distribution, it is speculated that seipin is a tissue-dependent and multi-functional protein. Studies using yeast and mammalian cells demonstrated that seipin plays a key role in adipogenesis, lipid droplet homeostasis and cellular triglyceride lipolysis (6–8). For example, seipin deficiency strongly impairs adipocyte homeostasis and leads to lipodystrophy (9). Knockdown of seipin in 3T3-L1 and C3H10T1/2 cell lines led to depression of terminal adipocyte differentiation (8). In addition, seipin deficiency alters lipid synthesis and lipid droplet homeostasis (10,11). However, the function of highly expressed seipin in the nervous system such as cortex, cerebellum, hippocampus and hypothalamus (2,12) remains to be elucidated.
Seipin is predicted to span the endoplasmic reticulum (ER) membrane twice with both N- and C-termini in the cytoplasm and a large luminal loop (12). Missense mutations (N88S and S90L) of seipin in the luminal loop activated unfold protein response to induce ER stress that led to the amyotrophy of hand muscles and the spasticity of lower limbs, termed seipinopathy (13,14). Moreover, the over-expression of human seipin mutants (N88S and S90L) in mice caused the death of alpha motor neurons in the spinal cord which was not associated with total loss of adipose tissue, severe insulin resistance, hypertriglyceridemia or fatty liver (14). In addition to lipodystrophy, another main characteristic phenotype of CGL2 patients was mental retardation (15) without the impairment of motor neurons (16). Therefore, it is of great interest to investigate the functional effects of seipin on locomotion and affective behaviors.
Seipin depletion has been reported to inhibit the expression of peroxisome proliferator-activated receptor gamma (PPARγ) and differentiation in 3T3L1 cells (8), murine embryonic fibroblasts and stromal vascular cells (17), mouse embryonic stem cells and primary preadipocytes in mice or human (18). Administration of the PPARγ agonist pioglitazone was able to rescue seipin-depletion induced inhibition of 3T3L1 cell differentiation (8). Low PPARγ levels had been reported in the brains of depressive patients (19). The Pro12Ala polymorphism in PPARγ2 was associated with a high risk for developing depression (20). Moreover, the activation of PPARγ effectively improved depression-like behaviors (21). Recently, knockdown of the seipin gene in cultured cortical neurons was reported to cause the reduction in excitatory post-synaptic currents via down-regulation of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors and cAMP response element-binding protein (22).
In the last 2 years, three independent models of systemic BSCL2/seipin-deficient mice (seipin-KO mice) have been published (9,17,23). All three models of seipin-KO mice mainly displayed a severe and consistent lipodystrophy with a dramatic loss of fat mass. Seipin-KO mice were also diabetic and developed severe hepatic steatosis. To determine the actual functions of seipin in neurons, we generated a mouse model in which the seipin was selectively depleted in the nervous system (seipin-nKO) using mice with the seipin gene flanked by LoxP crossed with mice overexpressing Cre transgene driven by a neuron-specific Nestin promoter (nestin-Cre). In this study, male and female seipin-nKO mice were used to investigate the influence of seipin deficiency on locomotion and affective behaviors. We further examined the expression of PPARγ in the hippocampus and cortex of seipin-nKO mice to explore molecular mechanisms underlying seipin-regulated locomotion and affective behaviors. Our results indicated that neuronal seipin deficiency, by reducing PPARγ expression, caused affective disorders in male mice that were rescued by E2-enhanced PPARγ expression.
Influence of seipin-nKO on spontaneous activity, affective behaviors and motor abilities
Spontaneous activity was initially examined using an open-field test (OFT). Male seipin-nKO mice had a tendency to perform less in travelling (Fig. 1A) and rearing (Fig. 1B), but, the group when compared with control mice failed to reach the significance (P > 0.05, n = 12). In contrast, neither the distance traveled nor the rearing numbers in female seipin-nKO mice had any difference from those in female control mice (P > 0.05, n = 12). The time spent in the central partition of the arena was significantly reduced in male seipin-nKO mice compared with controls (P < 0.01, n = 12 per group; Fig. 1C), but not in female seipin-nKO mice (P > 0.05, n = 12 per group).
Anxiety- and depression-like behaviors were further examined by elevated plus-maze (EPM), forced swim test (FST) and tail suspension test (TST). Consistent with the results of OFT, male, but not female, seipin-nKO mice spent less time in open arms of EPM than control mice (P < 0.01, n = 12; Fig. 1D). In addition, male seipin-nKO mice showed longer immobility during FST (Fig. 1E) and TST (Fig. 1F) than control mice (P < 0.01, n = 12). In contrast, compared with control mice, female seipin-nKO mice showed no difference in immobility during the FST or TST (P > 0.05, n = 12).
Motor coordination and abilities were examined by the rotarod test and swimming test. As shown in Figure 1G, when rotation speeds were at 4, 12, 20, 28 and 36 rpm, the time of male or female seipin-nKO mice stayed on the rotarod was nearly equal to that of controls (P > 0.05, n = 12). In addition, male and female seipin-nKO mice did not show a change in swimming speed compared with control mice (P > 0.05, n = 12; Fig. 1H). These results indicate that male seipin-nKO mice display anxiety- and depression-like behaviors without changes in spontaneous activity, motor coordination or locomotion abilities.
Involvement of E2 in gender-related affective disorders induced by seipin-nKO
In seipin-nKO mice, plasma levels of 17β-estradiol (E2) in females (Fig. 2A), testosterone in males (Fig. 2B) and corticosterone in either gender (Fig. 2C) did not significantly differ from controls (P > 0.05, n = 12). To determine whether E2 is involved in the gender-related affective disorders induced by the seipin deficiency, female seipin-nKO mice and control mice were ovariectomized (OVX) or male seipin-nKO mice and control mice were treated with E2 [120 μg/kg, subcutaneous (s.c.)] for 20 days. During the third week following OVX, the time spent in open arms of EPM was significantly decreased (P < 0.01, n = 12; Fig. 2D), while immobility during FST (P < 0.01, n = 12; Fig. 2E) and TST (P < 0.01, n = 12; Fig. 2F) was prolonged compared with sham-op female seipin-nKO mice. Although the OVX-control mice showed similar changes in either the time spent in open arms or the time of immobility in the FST and TST, the group comparison with sham-op control mice failed to reach the significance (P > 0.05, n = 12). In contrast, E2-treated male seipin-nKO mice, compared with vehicle-treated male seipin-nKO mice, spent more time in open arms of EPM (P < 0.01, n = 18; Fig. 2G) and less time immobile during FST (P < 0.01, n = 18; Fig. 2H) and TST (P < 0.01, n = 18; Fig. 2I). However, the treatment of male control mice with E2 (120 μg/kg, s.c.) for 20 days did not cause detectable changes in the time spent in open arms in EPM or immobile time during FST and TST (P > 0.05, n = 18). The results indicate that the loss of E2 action leads to the gender-related anxiety- and depression-like behaviors induced by seipin deficiency.
Influence of seipin-nKO on PPARγ expression and ER-stress
In accordance with reports by Payne et al. (18) and Fei et al. (25), PPARγ mRNA (Fig. 3A) and PPARγ protein (Fig. 3B) were decreased in the hippocampus and cortex of male seipin-nKO mice compared with controls (P < 0.01, n = 8). In contrast, no significant difference in PPARγ mRNA or PPARγ protein levels was found between female control mice and seipin-nKO mice (P > 0.05, n = 8). Interestingly, the levels of PPARγ mRNA and PPARγ protein in the hippocampus and cortex were elevated in male seipin-nKO mice by E2 treatment (120 μg/kg, s.c.) (P < 0.01, n = 8) and were reduced in OVX-female seipin-nKO mice (P < 0.01, n = 8). However, the well-established ER-stress markers immunoglobulin heavy-chain-binding protein (BIP; Fig. 3C) and C/EBP homologous protein (CHOP; Fig. 3D) in the hippocampus and cortex were not altered in male or female seipin-nKO mice (P > 0.05, n = 8). The results indicate that seipin deficiency depresses the expression of PPARγ, which can be recovered by E2 application.
Relation of reduced PPARγ with abnormal affective behaviors in male seipin-nKO mice
To determine whether low expression of PPARγ is related to affective disorders, male seipin-nKO mice were treated with the PPARγ agonist rosiglitazone [5 mg/kg, oral administration (p.o.)] for 30 days. As expected, the treatment of male seipin-nKO mice with rosiglitazone significantly increased the time spent in open arms of EPM (P < 0.01, n = 18; Fig. 4A) and reduced the immobility times in FST (P < 0.01, n = 18; Fig. 4B) and TST (P < 0.01, n = 18; Fig. 4C) compared with vehicle-treated male seipin-nKO mice. In contrast, the treatment of male control mice with rosiglitazone for 30 days did not cause the detectable changes in the time of open arms in EPM or the immobility time in FST and TST (P > 0.05, n = 18). To exclude the influence of rosiglitazone on the regulation of energy balance (24), we measured the body weight before and after the administration of rosiglitazone. The results showed that the treatment with rosiglitazone for 30 days did not cause the increase of body weight in male control mice or seipin-nKO mice compared with their basal levels (P > 0.05, n = 18; Fig. 4D). The results indicate that the seipin deficiency-induced down-regulation of PPARγ is responsible for anxiety- and depression-like behaviors.
Using adult seipin-nKO mice, the present study provides in vivo evidence that neuronal seipin deficiency, by decreasing PPARγ expression, causes affective disorders, as suggested by the following results. First, male seipin-nKO mice spent less time in the central field of OFT, less time in the open arms of EPM and more time immobile in FST and TST. Second, female seipin-nKO mice subjected to OVX displayed the same anxiety- and depression-like behaviors as male seipin-nKO mice. Third, treatment with either E2 or the PPARγ agonist rosiglitazone alleviated anxiety- and depression-like behaviors in male seipin-nKO mice. The main characteristic phenotype of CGL2 patients is severe lipodystrophy and mental retardation. Indeed, we observed that male 8-week-old systemic seipin knockout mice (seipin-sKO mice) showing severe fat loss (9) spent less time in open arms of EPM and longer immobile time during FST and TST than those in wild-type (WT) mice (Supplementary Material, Fig. S4A–C). The first question we should address may be whether the neurological symptoms are secondary to fat loss. We observed that the amount of adipose tissue in male seipin-nKO mice was not reduced compared with control mice (unpublished data), although their body weight was less than that in control mice. In addition, the treatment with rosiglitazone in male seipin-nKO mice alleviated the affective disorders, but it did not alter their body weight. Although OVX-female seipin-nKO mice appeared the anxiety- and depression-like behaviors, their body weight did not differ from sham-op seipin-nKO mice (data not shown). The results indicate that the affective disorders induced by the seipin deficiency are unlike to be secondary to lipodystrophy.
Seipin highly expressed in the hippocampus and cortex is likely to be implicated in mood disorders. Although the underlying mechanisms still remain largely unknown, in accordance with the in vitro results (25), the in vivo data in male seipin-nKO mice showed the down-regulation of PPARγ expression in the hippocampus and cortex. Additionally, seipin, as an exclusive ER-resident N-glycosylated protein, has been reported to play a role in generation of an endogenous ligand interfering with PPARγ (25). The down-regulation of PPARγ by seipin knockdown greatly impairs the differentiation of mouse embryonic stem cells (18). In the hippocampal dentate gyrus (DG) of mammal, stem cells retain the ability to proliferate and neural progenitor cells can further differentiate into neurons (26). The continual neurogenesis in adults is required to maintain hippocampal functions (27). Hippocampal neurogenesis has been associated with the reversal of depression (28). For example, anti-depressants can increase the number of newborn cells in the DG (29), while reduced cell proliferation attenuates the effects of monoaminergic anti-depressant (30). The activation of PPARγ has been reported to play an important role in controlling the proliferation of neural stem cells and their neuronal differentiation (31). The PPARγ agonist rosiglitazone or pioglitazone stimulates neural stem cells growth, while PPARγ-blockade causes cell death in a concentration-dependent manner via triggered caspase signaling (32). The stimulation of cell growth by PPARγ is associated with a rapid phosphorylation of extracellular signal-regulated kinases 1 and 2, and up-regulation of epidermal growth factor receptor and cyclin B. Additionally, PPARγ activation can enhance neurogenesis in the adult hippocampal DG (33) by decreasing neuronal nitric oxide synthase (34) or inhibiting the cytokine-activated NF-kB (35). Further efforts are necessary to determine whether the seipin deficiency-induced down-regulation of PPARγ affects neurogenesis in the hippocampal DG. On the other hand, the seipin protein has been reported to exert a potential role in enhancing axonal transport (14) and AMPA receptors expression (22). Therefore, it is proposed that the affective disorders in seipin-nKO mice may arise from deficits in neuronal function and imbalance of excitatory and inhibitory synaptic transmission.
An interesting finding in this study is that female seipin-nKO mice failed to develop affective disorders. Many researchers have paid close attention to the functional roles of E2 in affective disorders. E2 replacement therapy is beneficial to women who suffer from postpartum and perimenopausal affective disorders. Our results showed that the treatment of male seipin-nKO mice with E2 could alleviate the anxiety- and depression-like behaviors and elevate the levels of PPARγ mRNA and protein. In contrast, OVX-female seipin-nKO mice displayed affective disorders accompanied by the decline of PPARγ level. In addition, we observed that female 8-week-old seipin-sKO mice showed the anxiety- and depression-like behaviors (Supplementary Material, Fig. S5A–C) with the decline of PPARγ level compared with WT mice (Supplementary Material, Fig. S5D). More importantly, the levels of plasma E2 in female seipin-sKO mice were lower than that in WT mice (Supplementary Material, Fig. S5E) probably owing to ovarian seipin-deficiency (Supplementary Material, Fig. S5F). The activation of intracellular estrogen receptors β has been reported to enhance PPARγ expression (36). The expression of PPARγ depends on E2 action, but PPARγ agonists can normalize PPARγ expression with a lack of E2, implicating a regulative crosstalk between PPARγ and E2 for controlling PPARγ expression (37). However, our results showed that the deprivation of E2 in OVX-female control mice did not affect the PPARγ expression and affective behaviors. In male control mice treated with E2, the PPARγ level was not elevated. Although elucidation of the underlying mechanism awaits further studies, it is conceivable that the presence of E2 is able to correct the down-regulation of PPARγ induced by seipin deficiency to prevent abnormal affective behaviors. On the other hand, exposure to E2 is able to promote cell proliferation in male rats (38) and the survival of newly generated neurons in male meadow voles (39). Taking our data and other studies into account, we propose that E2 can protect hippocampal neurogenesis against the PPARγ deficiency-induced impairment in female seipin-nKO mice.
Another important finding in this study is that male and female seipin-nKO mice failed to display obvious locomotion deficits. Although male seipin-nKO mice showed a decreased tendency in traveling distance and rearing behavior during OFT, motor coordination on the rotarod and swimming speed in water tank were not significantly different from control mice. In particular, neither spontaneous activity nor locomotion ability in female seipin-nKO mice was different from that in control mice. In comparison with control mice, the structure of the spinal cord and the number of alpha motor neurons in the spinal cord were not altered in male or female seipin-nKO mice (Supplementary Material, Fig. S3). Because male seipin-nKO mice showed more anxiety and depression than control mice, it is conceivable that the slightly reduced spontaneous locomotion is likely due to mood disorders rather than deficits in motor function. Consistent with our results obtained from seipin-nKO mice, GCL2 patients showing mental retardation do not develop the locomotion deficits and the impairment of motor neurons (13). Previous studies reported that N88S/S90L mutations resulted in formation of cytoplasmic inclusions and enhanced ubiquitination, leading to unfolded protein accumulation in the ER and eventually ER-stress-associated cell death (12,40). A recent study has reported that human seipin over-expression led to the death of alpha motor neurons in the spinal cord by enhancing the activation of the autophagy pathway rather than inducing ER stress (40). Indeed, the protein levels of the ER-stress markers BIP and CHOP were not increased in male and female seipin-nKO mice. Therefore, this discord may arise from the difference in the experimental models.
In summary, the results obtained from seipin-nKO mice will bring new insights into physiological function of seipin in regulation of affective behaviors. Although much more work needs to be performed, the seipin deficiency-induced affective disorder that is implicated in CGL2 patients, at least partly, relies on the decline of the PPARγ-mediated signaling pathway.
MATERIALS AND METHODS
The present study was approved by the Animal Care and Ethical Committee of Nanjing Medical University. All animal-handling procedures followed the guidelines of the Institute for Laboratory Animal Research of the Nanjing Medical University. The mice were maintained in a constant environment (temperature 23 ± 2°C, humidity 55 ± 5% and 12 : 12 h light/dark cycle) in the animal research center of Nanjing University. They received a standard laboratory chow diet before and after all procedures.
Generation and genotype identification of seipin gene knockout mice
Neuron-specific seipin knockout (seipin-nKO) mice
Seipin loxP/loxP mice were generated as described elsewhere (9). Neuron-specific deletion of seipin exon 3 was induced by crossing mice with the loxP seipin allele to transgenic mice expressing Cre recombinase driven by a neuron-specific promoter (nestin-Cre; Jackson Laboratory, USA). Progenies were screened by PCR for loss of the seipin exon 3 and presence of nestin-Cre. The genotype of seipin-nKO mice was identified by PCR using genomic DNA from their tail (Supplementary Material, Fig. S1A). The DNA was amplified with the following primers: 5′-CTTGTCTCAAAGGGGTCT-3′ (forward primer for loxP) and 5′-TCAACAGAACAGACGCT-3′ (reverse primer for loxP); 5′-GCGGTCTGGCAGTAAAAACTATC-3′ (forward primer for nestin-Cre) and 5′-GTGAAACAGCATTGCTGTCACTT-3′ (reverse primer for nestin-Cre). By analysis of reverse transcription-polymerase chain reaction (RT-PCR), the lack of seipin was determined in the brain (cortex and hippocampus) of seipin-nKO mice (Supplementary Material, Fig. S1B and C). Systemic seipin knockout (seipin-sKO) mice: the seipin-sKO mice were produced as described elsewhere (9). The genotype identification of seipin-sKO mice was performed according to this publication.
The male and female seipin-nKO mice appeared grossly normal from birth compared with controls. The reduction in body weight was observed in 8-week-old male and female seipin-nKO mice (P < 0.05, n = 12; Supplementary Material, Fig. S2A) without the decrease in the amount of adipose tissue (unpublished data). In addition, the size and shape of external brain had no significant difference between 8-week-old control mice and seipin-nKO mice (Supplementary Material, Fig. S2B). No major alteration of hippocampal size and shape in 8-week-old seipin-nKO mice was found compared with control mice, as illustrated in Supplementary Material, Figure S2C. The thickness of pyramidal cell layer in the CA1 region or granular cell layer in DG, and aspect of neuronal cells were not altered in male and female seipin-nKO mice compared with control mice. Eight-week-old seipin-nKO mice and seipin loxP/loxP mice (as control mice), seipin-sKO mice and WT littermates (WT mice) were used at the beginning of all experiments.
A single cohort of animals was used for the following test sequence: open-field test → elevated plus-maze → forced swim test → tail suspension test → swimming tank → rotarod test. Tests were spaced by at least 2 weeks, and the order of testing was chosen such that tests involving lower stress levels (open-field test and elevated plus-maze) preceded those involving higher stress levels (forced swim test and tail suspension test). All behavioral performance data were captured by video-monitor (Winfast PVR; Leadtek Research Inc., Fremont, CA, USA) and analyzed using TopScan Lite 2.0 (Clever Sys., Reston, VA, USA).
Mice were examined in a cuboid plexiglass box (60 cm × 60 cm × 40 cm) with the gray floor divided into 16 equal squares. The central zone was defined as the central 4 squares (30 cm × 30 cm). Each mouse was placed in a corner of the arena and allowed to freely explore for 6 min. The following parameters were evaluated: (i) total distance travelled (mm/6 min) was calculated as number of partitions crossed within 6 min; (ii) rearing frequency within 6 min; (iii) the time spent in the center region of the arena.
The apparatus consisted of two open arms (23.5 cm × 8 cm), painted white, and two enclosed arms (23.5 cm × 8 cm × 20 cm high), painted black. The maze was raised to a height of 38.5 cm above the floor. Mice were placed in the center area facing one of the open arms and their movement and time spent in the different arms were analyzed for 6 min.
Forced swim test
Each mouse was placed in a glass cylinder (300 mm high, 280 mm in diameter) filled with water to a height of 20 cm (25 ± 1°C). Mice were forced to swim for 15 min and subjected to 6 min swimming test 24 h later. Total immobility time (minimal movements to keep the head above water) was calculated.
Tail suspension test
Mice were suspended by the tail using adhesive tape to a rod 60 cm above the floor as described previously (41). The trials were conducted for 6 min, during which the duration of immobility was recorded.
To monitor swimming movement, mice were trained to swim from one end of a water-filled glass tank to a visible escape platform at the opposite end (42). The glass tank was 100 cm long and 6 cm wide and was filled with water (25 ± 1°C) to a depth of 20 cm. The visible escape platform was made from black plexiglass (6 cm square and 20.5 cm high) with the top surface 0.5 cm above the water level. During the training period, each mouse was placed at one end of the tank and learned to swim straight to the visible escape platform at the opposite end. After training for 3 days, mice were given two trials per day for three consecutive days, during which their swimming speeds were recorded.
The rotarod apparatus (Accelerating Model, Biological Research Apparatus, Varese, Italy) was used to measure forelimb and hindlimb motor coordination and balance. All mice were trained for at least 3 days until they were able to stay on the rotarod for 60 s at 16 rpm. Mice were then given two trials at 4, 12, 20, 28 and 36 rpm for 60 s each. The mean latency to fall off from the rotarod was recorded.
Ovariectomy (OVX) and drug treatment
Some female mice were bilaterally ovariectomized (OVX) as described previously (43). E2 (Sigma-Aldrich, St Louis, MO, USA) and the PPARγ agonist rosiglitazone (Enzo, Farmingdale, NY, USA) were dissolved in dimethyl sulfoxide (DMSO) and then diluted in 0.9% saline to a final concentration of 0.5% DMSO. Some male mice were treated daily with the s.c. injection of E2 (120 μg/kg) (44) or p.o. of rosiglitazone (5 mg/kg) (45).
Measurement of E2, testosterone and corticosterone
To measure E2, testosterone and corticosterone, blood samples were taken at 0900–1000 by jugular venipuncture in male and female seipin-nKO mice. Plasma (300 μl per mouse) was separated by centrifugation at 4°C for 15 min at 3500 rpm (Thermo Scientific, Waltham, MA, USA) and stored at −80°C until the assay. Levels of testosterone and E2 were measured using a RIA kit provided by the National Hormone and Peptide Program (Baltimore, MD, USA). The intra- and inter-assay coefficients of variation were 6.2 and 7.4% for testosterone, and 6.0 and 5.8% for E2. Corticosterone levels were measured with a corticosterone ELISA kit according to the instructions of the manufacturer (Cayman Chemical, Ann Arbor, MI, USA).
Animals were deeply anesthetized with chloral hydrate (400 mg/kg, i.p.) and cardinally perfused with cold PBS followed by 4% paraformaldehyde. Brains and the L3–L5 region of spinal cord were removed and processed for paraffin embedding. Coronal sections (5 μm in thickness) were placed in gelatine-coated slides, and then were stained with toluidine blue. Hippocampus and spinal cord were observed using a conventional light microscope (Olympus DP70) with 10× objective. For spinal motor neuron quantitation, every sixth section (5 μm in thickness) was obtained for consecutive analysis of cell quantification. The slides were coded before analysis. The experimenter was blind to all samples until they were counted. The cells with nuclei visible and diameter > 25 μm in ventral horn (VH) region of spinal cord were counted. The VH region with an area of greater than 600 μm2 was analyzed. The number of spinal motor neurons counted in 20 sections of the L3–L5 region was averaged to obtain a single value per VH region.
Western blot analysis
Tissues obtained from the hippocampus and cortex were homogenized in 1% Nonidet P-40 lysis buffer containing 50 mm Tris, 150 mm NaCl, 0.02% NaN3, and complete protease inhibitors (Roche, Indianapolis, IN, USA). The homogenates were centrifuged for 15 min at 12 000 rpm (Thermo Scientific), and the supernatants were collected. Proteins (50 μg) were loaded in each lane for separation in sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS–PAGE) and transferred onto nitrocellulose membranes. Blotting membranes were incubated with blocking solution (5% non-fat dried milk) for 1 h at room temperature, washed three times and then incubated with a rabbit monoclonal anti-PPARγ antibody (1:1000, Cell Signaling, Beverly, MA, USA), a rabbit monoclonal anti-BIP antibody (1:1000, Bioworld Technology, Minnesota, MN, USA) or a mouse monoclonal anti-CHOP antibody (1:1000, Cell Signaling) at 4°C overnight. The internal control was performed using a GAPDH antibody (1:5000, Bioworld Technology) or β-actin antibody (1:2000, Cell Signaling). After being washed with TBST, the membranes were incubated for 1 h with HRP-labeled secondary antibody (1:10000, Millipore, Billerica, MA, USA) followed by an enhanced chemiluminescence system (ECL, Amersham Biosciences, Piscataway, NJ, USA), and autoradio-graphic exposure to Hyperfilm ECL (Amersham Biosciences). Signal quantification was carried out using Quantity One image software (Bio-Rad).
Reverse transcription-polymerase chain reaction (RT-PCR)
Real-time PCR was performed as described previously (46). Total RNA was isolated from the hippocampus and cortex with TRIzol reagent (Invitrogen, Camarillo, CA, USA) and reverse-transcribed into cDNA using a Prime Script RT reagent kit (Takara, China) for quantitative PCR (ABI Step One Plus, Foster City, CA, USA) in the presence of a fluorescent dye (SYBR Green I; Takara). The relative expression of genes was determined using the 2-ΔΔct method with normalization to GAPDH expression. The primers used for seipin were 5′-GGCTCCTTCTACTACTCCTACA-3′ and 5′-CCGATCACGTCCACTCTT-3′; the primers for PPARγ were 5′-GCTTATTTATGATAGGTGTGATC-3′ and 5′-GCATTGTGAGACATCCCCAC-3′; the primers for GAPDH were 5′-ACCACAGTCCATGCCATCAC-3′ and 5′-ACCACAGTCCATGCCATCAC-3′.
Data were retrieved and processed with PulseFit (HEKA Elektronik) and Microcal Origin 6.1 software. Group data are expressed as the means ± standard errors (SEMs). Difference between two groups was evaluated using Student's t-test. When analyzing one-variable experiments with more than two groups, statistical significance was evaluated using ANOVA followed by Bonferroni's post hoc tests. Statistical analysis was performed using Stata 7 software (STATA Corporation, College Station, TX). Differences at P < 0.05 were considered statistically significant.
Conflict of Interest statement. None declared.
This work was supported by grants for 973 (No. 2014CB943303), NSFC (No. 81361120247; 31171440) to L.C.; 973 (No. 2011CB503900) and NSFC (No. 81121061) to G.L.