Calcineurin (CN) is a unique phosphatase that is regulated by the second messenger Ca2+ together with calmodulin. CN is a heterodimer composed of a catalytic subunit called calcineurin A (CNA, ∼61 kD) and a regulatory subunit (CNB, ∼19 kD) (1). CN is widely distributed among eukaryotes, with a structure that is conserved from yeast to humans and exhibits an extremely high level of expression in the mammalian brain. However, it has a narrower range of substrates compared with other phosphatases. CN dephosphorylates important neuronal substrates, including cytoskeletal proteins, such as tau, which is phosphorylated abnormally at multiple Ser/Thr sites in early Alzheimer's disease (AD) (2). As CN has important roles that are dependent on its activity in signal transduction, many studies exploring the biological roles of CN have focused on its regulation.
It has been shown that the phosphatase activity of CN is sensitive to oxidative stress and may be modulated by the intracellular redox potential (3). Cu, Zn superoxide dismutase (SOD1) has been identified as a factor that prevents inactivation of CN in vitro. A SOD1-deficient yeast strain had very low CN activity and presented a phenotype resembling the effects of CN deletion, such as the inability to recover from G1 arrest after pheromone treatment and lack of adaptation to high-salt stress (4). Moreover, overexpression of the wild-type, antioxidant SOD1 in SY5Y cells promotes CN activity and protects the phosphatase from inactivation (5).
Regulators of calcineurin (RCANs) are a novel family of CN endogenous regulators that have been shown recently to modulate CN activity under physiological and pathological conditions (6). RCAN1 was first identified as a Down's syndrome critical region-localised gene on human chromosome 21. The RCAN1 gene contains seven exons, and differential promoter usage and first exon choice can generate several transcripts. The different mRNAs contain one of the four possible initial exons (E1–E4) in addition to the three exons (E5–E7) that are common to all isoforms of RCAN1 mRNA. All mRNA isoforms contain exons 5–7, whereas the three most studied isoforms also contain either 29 amino acids (RCAN1-1S), 55 amino acids (RCAN1-1L), which are encoded by exon 1, or 29 amino acids (RCAN1-4) encoded by exon 4 (7).
RCAN1 proteins are expressed in multiple tissues but primarily in the brain and cardiac muscle. The first indication that RCAN1 might play a central role in pathological processes was the finding that mammalian cells increase RCAN1-4 expression after exposure to damaging stimuli associated with Ca2+ influx, such as oxidative stress (8, 9). A critical breakthrough came when it was demonstrated that RCAN1 proteins can bind to CN and inhibit its activity in mouse and human cells (10–12).
The primary CN-binding portion of RCAN1 is encoded by exon 7, which is common to all RCAN1 isoforms. Interest in RCAN1 was reinforced by the discoveries that the RCAN1-1L transcript and protein both increase in the brains of individuals with AD (13–16). As RCAN1 is a CN inhibitor, and CN represents a large portion of total brain protein, alterations in expression of RCAN1 would be expected to have an important effect in the brain. In fact, CN activity levels have been reported to decrease in AD (2). CN inhibition has been associated with tau phosphorylation at threonine 181 and 231, which is observed in paired helical filament preparations from AD brains. Therefore, it has been proposed that long-term induction of RCAN1-1L expression might be associated with neurodegeneration in AD.
Cyclosporine A (CsA) is an important immunosuppressive drug that is used in solid organ transplants (17). It is well known that CsA exerts its immunosuppressive effect by inhibiting CN in vitro and in vivo. CsA binds to a distinct family of endogenous receptor proteins, collectively called cyclophilin (CyP), forming CsA–CyP complexes that potently and selectively inhibit CN through a gain-of-function mechanism.
Here, we have investigated RCAN1 and SOD1 expression in mouse brain after an oral dose of CsA. We first demonstrate that there was a significant up-regulation of RCAN1-1L and RCAN1-4 protein isoforms in mouse brain after chronic oral administration of a 20 mg/kg/day dose of CsA. Excessive RCAN1 proteins resulted in more potent CN inhibition compared with the previously reported data (18). In SY5Y cells, however, administration of CsA resulted in reduction of SOD1 and did not alter RCAN1-1 mRNA expression. It is our aim to evaluate the modulation of CN activity in mouse brain by chronic oral administration of CsA.
Materials and Methods
RII peptide, a CN substrate, was obtained from BioMoL Research Laboratories (Plymouth Meeting, PA). CsA and okadaic acid were purchased from Sigma Chemical (St. Louis, MO). RCAN1 mouse antihuman monoclonal antibody was obtained from LifeSpan Biosciences (Seattle, WA) and rabbit polyclonal to SOD1 was obtained from Abcam (Hong Kong). An antibody to tubulin (β-4) was purchased from ProteinTech Group (Chicago, IL), and an antibody to CNA was prepared by our lab. Goat antimouse and antirabbit peroxidase-conjugated secondary antibodies were obtained from Jackson ImmunoResearch Laboratories (West Grove, PA). Chemiluminescence assay kits were purchased from Pierce Chemical (Rockford, IL). All other reagents were of standard laboratory grade and the highest quality was available from commercial suppliers. Male Kunming mice (weight, 16 ± 2 g; 4 weeks old) were supplied by the Experimental Animal Centre of Peking University.
Preparation of Mouse Tissue Extracts
The animals were group-housed under the following laboratory conditions: temperature of 20 ± 1 °C, humidity of 40–60%, and 12:12-h light/dark cycle. Mice had free access to food and water throughout the course of the experiments. The animals were treated in accordance with the current law and the NIH Guide for Care and Use of Laboratory Animals. For the in vivo studies, CsA was suspended in a 0.2% carboxymethyl cellulose solution and administered intragastrically at a concentration of 20 mg/kg/day. The oral volume was kept constant at 20 mL/kg. After a 25-day oral administration of CsA, the mice were sacrificed. The whole brain was immediately removed and homogenised by passing through a 16G syringe needle many times to break up cells at 4 °C into a solution of 50 mM Tris-HCl, pH 7.5, 0.1 mM ethylene diamine tetraacetic acid (EDTA), 0.1 mM ethylene glycol tetraacetic acid (EGTA), 1.0 mM dithiothreitol (DTT), 0.2% (v/v) NP-40, 1.0 mM phenylmethylsulphonyl fluoride, 5 μg/mL leupeptin, 5 μg/mL aprotinin, and 2 μg/mL pepstatin. Air bubbles were avoided during this stage of the preparation. The tissue homogenates were then centrifuged at 16,000 × g, and the supernatants were used to assay the phosphatase activity and for Western blot analysis.
Human neuroblastoma cells (SY5Y) were cultured in a 1:1 (v/v) mixture of Dulbecco's minimum essential medium with Ham's F12 medium (GIBCO-BRL, Life Technologies, Grand Island, NY), containing nonessential amino acids, 10% fetal calf serum, 100 units of penicillin per millilitre, and 100 μg of streptomycin per millilitre. The cells were incubated at 37 °C in humidified air containing 5% CO2. Mitochondrial respiration, an indicator of cell survival rate, was assessed by the mitochondrial-dependent reduction of 3-(4,5-dimethylthiazol-2-yl)-2,5 -diphenyltetrazolium bromide (MTT) to formazan, as described previously (19). CsA was dissolved in DMSO to generate 15-mM stock solutions. The cells were cultured in either the presence or the absence of CsA at 37 °C for 1 h. The cells were washed twice with 1 mL of phosphate-buffered saline on ice and lysed in 40 μL of hypotonic buffer per millilitre of lysis solution. The hypotonic buffer contained 50 mM Tris-HCl, pH 7.5, 1 mM EDTA, 0.1 mM EGTA, 1.0 mM DTT, 0.2% NP-40, 1.0 mM phenylmethylsulphonyl fluoride, 5 μg/mL leupeptin, 5 μg/mL aprotinin, and 2 μg/mL pepstatin. The lysates were subjected to three cycles of freezing in liquid nitrogen followed by thawing at 30 °C and then centrifuged at 4 °C for 10 min at 12,000 × g. The supernatants were used to assay the phosphatase activity and for Western blot analysis. The protein concentration was determined by the Bradford method.
CN activity in tissues was determined by the Calcineurin Cellular Assay Kit PLUS-AK-816 (BioMol) according to the manufacturer's instructions. The kit was a product in the BIOMOL GREEN™ QuantiZyme™ Assay system and was a complete colorimetric assay kit for measuring cellular CN phosphatase activity. Extracts of tissues were prepared, and CN activity was measured as the dephosphorylation rate of a synthetic phosphopeptide substrate (RII peptide). The amount of PO43− released was determined colorimetrically with the BIOMOL GREEN reagent. Phosphatase activities were presented in the form of picomoles of phosphate released per milligram of protein per minute.
RNA Isolation and Real-Time PCR
RNA was isolated from the whole-brain tissue and SY5Y cells using the RNAprep pure Tissue Kit (Tiangen, Beijing, China). Then, the corresponding cDNA was synthesised using 1 μg of total RNA, M-MLV reverse transcriptase (TaKaRa, Dalian, China), and Oligo (dT) eight Primers (TaKaRa, Dalian, China).
Real-time quantitative PCR (qRT-PCR) amplification of RCAN1-1L transcripts in mouse brain was accomplished using forward and reverse primers corresponding to sequences within exons 1 and 5, respectively. The forward primer sequence used was 5′-ggtgatgtccttgtcatacgtc-3′, and the reverse primer sequence was 5′-atcgcctgtcacctggac-3′. The sequence for the RCAN1-1L transcript was extracted from the GenBank database (Accession # NM-004414). One microlitre of reverse transcription product was amplified in an ABI 7500 Fast Real-Time PCR machine with 250 pM of each primer and 12.5 μL of the 2×TranStart™ Green qPCR SuperMix (TransGenBiotech, China) in a 20-μL reaction volume under the following conditions: 5 sec at 95 °C, 15 sec at 60 °C, and then 10 sec at 72 °C, for 40 cycles. In duplicate reactions, the PCR was terminated during the exponential phase, and PCR products were resolved on a 0.8% agarose gel in Tris–acetate–EDTA buffer after conventional procedures. To normalise between samples, β-actin was used as a control. The forward primer used for this portion of the experiment was 5′-gtgacagcagtcggttggag-3′ and the reverse primer was 5′-agtggggtggcttttaggat-3′.
Semi-quantitative PCR amplification of the RCAN1-1 transcript in SY5Y cells was achieved using the following set of forward and reverse PCR primers: (5′-gactggagcttcattgactgcgaga-3′) and (5′-accacgctgggag tggtgtcagtcg-3′), respectively. To normalise gene expression levels, the following two primers specific to GAPDH cDNA were used: 5′-agaaggctggggctcatttg-3′ (forward primer) and 5′-aggggccatcca catcttc-3′ (reverse primer). Semi-quantitative PCR was performed using TransTaq DNA Polymerase High Fidelity (TRANS). To compare the abundance of RCAN1-1 transcripts in control and CsA-treated SY5Y cells, RCAN1-1 expression was calculated relative to that of GAPDH transcripts.
Expression of RCAN1, SOD1, CNA, and tubulin proteins was determined by Western blot, as described elsewhere (20). RCAN1 mouse antihuman monoclonal antibody was purchased from LifeSpan Biosciences and its immunogen was a recombinant protein corresponding to amino acids 1–100 of human RCAN1. The protein blots were developed with anti-RCAN1 (1:100), anti-SOD1 (1:1,000), anti-CNA (1:2,000), or antitubulin (1:800) antibodies and visualised with an enhanced chemiluminescence assay kit (Pierce, Rockford, IL), followed by exposure to Kodak X-Omat BT film. The immunoreactivity of the bands was analysed quantitatively with BIO-RAD Quantity One software and expressed as the sum of the optical density relative to the control.
The data were expressed as the mean ± S.D. and analysed with SPSS 16.0 statistical software. The one-way ANOVA, followed by LSD post hoc tests, was used to determine the statistical significance of differences of the means. For a single comparison, the significance of differences between the means was determined by a t-test.
RCAN1 Proteins and RCAN1-1L mRNA are Both Up-regulated in Chronic CsA-Treated Mouse Brain
First, RCAN1 expression in chronic CsA-treated mouse brain was assessed. Previous study had shown that the RCAN1 gene was expressed primarily in mouse and human brain tissues, and the RCAN1 protein isoforms most studied in brain tissue were RCAN1-1L, RCAN1-1S, and RCAN1-4. All RCAN1 isoforms differ in their initial exon but share the same 168 amino acids encoded by exons 5, 6, and 7 (11). To determine which of the, if any, protein isoforms were expressed, mouse brain tissue extracts were prepared for Western blot. The antibody purchased from LifeSpan Biosciences was raised against amino acids 1–100 of RCAN1, which is a portion of the RCAN1 N-terminus that comprises exon 1 and part of exon 5. Exon 5 encodes the common region to all predicted isoforms, and thus, the antibody should recognise all forms of RCAN1 protein in the brain. Accordingly, a minor band of approximately 29 kD, representing RCAN1-4, and a 38-kD band, representing RCAN1-1L, were detected. A band for RCAN1-1S was also observed at approximately 31 kD but was very weak and difficult to detect and quantify (data not shown). In good agreement with a previous study of mouse brain, RCAN1-1L was much more abundant than RCAN1-4 (21). We found that initially, oral administration of 20 mg/kg CsA for 7 days had no significant effect on RCAN1 expression (data not shown), but compared with control brains, there was a significant up-regulation of RCAN1-1L and RCAN1-4 protein isoforms in mouse brain after 25 days. The expression of RCAN1-1L increased about 1.5-fold (Fig. 1A).
We then compared RCAN1-1L mRNA expression in brain tissue from CsA-treated and control samples by qRT-PCR amplification of cDNA. Our results showed a clear up-regulation of RCAN1-1L mRNA in CsA-treated mouse brain 25 days after oral administration of CsA (Fig. 1B).
SOD1 Expression is Decreased in Chronic CsA-Treated Mouse Brain
SOD1 is a ubiquitously expressed homodimeric cytosolic enzyme that dismutates superoxide radical into H2O2 and O2 and normally protects CN from inactivation. SOD1 expression in mouse brain was determined by Western blot analysis using an anti-SOD1 polyclonal antibody. After 25 days, chronic CsA-treated mice showed decreased levels of SOD1 expression relative to those of the control group (Fig. 2). Under our experimental conditions, SOD1 protein levels in CsA-treated mouse brain were approximately 33% lower than those in the control group.
Chronic Administration of CsA Produces Potent Inhibition of CN Activity Without Altering the Amount of CN in Mouse Brain
The CN cellular assay kit employs a convenient 96-well microtitre plate format with all reagents necessary for measuring CN activity in tissue/cellular extracts. We assayed CN activity and the total levels of the catalytic subunit of CN (CNA) in brain extracts from CsA-treated mice and the control group.
We calculated the CN activity in normal mouse brain and found it to be approximately 6,269 pmol/min/mg protein (25 days). In comparison, CN activity in CsA-treated mouse brain was only approximately 2,046 pmol/min/mg protein (Fig. 3A). To ensure that the differences in CN activity in brain extracts from the two groups were accurate, we assayed CNA protein content by Western blot using a polyclonal antibody made by our lab. Figure 3B shows that the amount of CNA protein was not significantly affected. Furthermore, treatment with CsA for 25 days reduced CN activity by 68%.
CsA-Treated Mice Display Overexpression of RCAN1-4 and Inhibition of CN Activity in Kidney and Spleen Tissues
To compare the sensitivity of mouse tissues to CN inhibition by CsA, we also assayed CN phosphatase activity in the kidney and spleen. These CN activities were intermediate and comparable to those previously reported (18). As calculated, CsA inhibited 54 and 55% of CN phosphatase activity in the kidney and spleen, respectively (Fig. 4A). Thus, CN activity was clearly highest in the brain, at levels 5- to 10-fold higher than in other tissues.
We next examined RCAN1 protein expression in kidney and spleen tissues after oral administration of 20 mg/kg/day CsA for 25 days. RCAN1-4 protein (∼29 kD) in the CsA-treated group was expressed at a higher level than in the control group for these tissues (Fig. 4B). However, using this antibody, we did not detect RCAN1-1L and RCAN1-1S in either the kidney or the spleen. The difference in RCAN1 basal expression between mouse brain and kidney may be partially attributable to differences in CsA-induced CN inhibition in these tissues.
CsA Inhibits CN Activity and Decreased SOD1 Expression but has no Significant Effect on RCAN1-1 mRNA Expression in SY5Y Cells
After treatment of SY5Y cells with different concentrations of CsA for 1 h at 37 °C, we measured cell viability using the MTT assay. Compared with the vehicle control, no difference in cell viability was observed upon 40 nM, 1 μM, or 25 μM CsA treatment (data not shown). Therefore, we used these concentrations of CsA for the rest of the studies in SY5Y cells.
We assayed CN activity and RCAN1-1 mRNA expression (including RCAN1-1L mRNA and RCAN1-1S mRNA) in SY5Y cell extracts. As expected, treatment with CsA dramatically decreased CN activity by 30, 75, and 82% with respect to the concentrations listed above (Fig. 5A). These results indicated that short-term CsA treatment can affect CN activity through CsA–Cyp complex formation. We then examined RCAN1-1 mRNA levels by RT-PCR and found that the mRNA expression levels were similar in extracts from CsA-treated and control cells (Fig. 5B). Nonetheless, we examined the amount of RCAN1 protein by Western blot in SY5Y cells; however, no bands could be detected in the presence or the absence of CsA. Thus, RCAN1 protein expression may be very low and therefore undetectable in SY5Y cells. In addition, in these in vitro studies, the cells were exposed to CsA for 1 h, a time period that is most likely too short to elicit obvious changes in RCAN1 protein expression.
We then examined the effect of CsA on SOD1 protein expression in SY5Y cells. Compared with the control group, SOD1 protein expression was significantly reduced by CsA treatment. The inhibition of SOD1 expression was found to be dose dependent (Fig. 5C). Increasing doses of CsA reduced SOD1 expression to 56, 27, and 13% of the control level.
Our results showed that long-term treatment with 20 mg/kg/day CsA resulted in potent CN inhibition in mouse brain. The significant reduction of CN activity in vivo resulted partially from increased expression of RCAN1 and partially from decreased expression of SOD1. In the present in vitro studies, we found that short-term CsA treatment did not affect RCAN1-1 mRNA expression in SY5Y cells.
In the previous study, the distribution of CsA into various tissue compartments from 1 to 72 h after oral administration of 100 mg/kg CsA was determined (18). CsA accumulated preferentially in the kidney, followed by the spleen, heart, whole blood, testis, and brain. Inhibition of CN activity in these tissues was correlated with the concentration of CsA. Brain tissue was highly resistant to inhibition by CsA because physiological barriers, such as the blood–brain barrier (BBB), and the abundance of CsA-binding proteins, such as immunophilins, determined the degree to which CsA accumulated in each tissue in vivo. CsA was primarily taken up from lipoprotein at the blood–brain interface, but because of the tight junctions at the blood–brain and blood–CSF barriers, CsA effectively became trapped in cerebral endothelial cells and the choroid plexus (22). In addition, Dohgu et al. (23) investigated the mechanism by which CsA causes BBB dysfunction in rodent brain endothelial cells. CsA inhibits the adenylyl cyclase/cyclic AMP/PKA signalling pathway activated by adrenomedullin, leading to the impairment of the brain endothelial barrier function. Therefore, CsA cannot significantly cross the BBB. For example, no CsA-related compounds were detected in the brain or spinal cord tissue of seven transplant patients who received CsA therapy (24).
However, we found that CN activity was inhibited in the brain after chronic oral administration of CsA. We first investigated RCAN1 expression in long-term CsA-treated mouse brain. After 25 days of administration, significant up-regulation of RCAN1 protein was observed, a major finding of our study. Using qRT-PCR, we found that RCAN1-1L mRNA expression was also increased and is thus consistent with the Western blot results. There were ample data indicating that RCAN1 protein can bind CN and that increased RCAN1 levels can inhibit the phosphatase activity of CN protein (25). The three-dimensional structure of the CN–CsA–CyP complex indicates that the CyP–CsA complex binds to a composite surface formed by CNA and CNB and lies over the active site of CNA. RCAN1–CN recognition is potentially multivalent, however, with CN interaction sites at the amino- and carboxyl-terminal domains (26). We suspected that overexpression of RCAN1 may further contribute to CN inhibition in brain tissue, resulting in a more potent reduction in CN activity.
It has been shown that CN phosphatase activity is also sensitive to oxidative stress and may be modulated by the intracellular redox potential (27, 28). Stemmer reported that a stabilising factor in crude bovine brain extracts protected CN from inactivation. This factor was subsequently isolated and identified by Wang et al. (4) as the superoxide anion scavenger, SOD. These authors also reported that although SOD protected CN from inactivation, superoxide anions promoted its inactivation. It has been reported that high concentrations of CsA induce oxidative stress in rat kidneys through both xanthine oxidase activation and impairment of the antioxidant defence system, which accelerates oxidation reactions in kidney tissue (29). In our study, we observed CsA-induced impairment of antioxidant defence system in mouse brain and SY5Y cells, and decreased SOD1 protein expression was also thought to contribute to CN inhibition.
RCAN1 was discovered as an oxidative stress gene and a calcium stress response gene that can protect cells from various forms of stress (8). Exposure of primary neurons to H2O2 induced RCAN1 mRNA expression after treatment. Western blot analysis revealed that the amount of RCAN1 decreased notably between 4 and 5 h after exposure to H2O2 due to protein degradation (30). It is obvious that the RCAN1 protein content is determined by the balance of RCAN1 synthesis versus degradation. Therefore, the elevated RCAN1 protein concentration in CsA-treated mouse brain may be due to both the up-regulation of RCAN1 mRNA expression and the down-regulation of RCAN1 protein degradation. It has been elucidated recently that the degradation of RCAN1 is mediated by both the chaperone-mediated autophagy pathway and the ubiquitin proteasome pathway. Dysfunctions of these pathways may contribute to the accumulation of RCAN1 protein in mouse brain (31).
It is worth noting that RCAN1 expression is required for some, but not all, cellular pathways of CN-dependent signal transduction. Evidence indicates that RCAN1 expression in Saccharomyces cerevisiae is CN dependent, but unfortunately, a role for CN in the induction of RCAN1 gene expression in higher eukaryotes has not been investigated (32). Future studies are warranted to further investigate the connection between RCAN1 expression and CN activity in vivo. Taken together, our results suggest that RCAN1, SOD1, and CN may influence one another and play an important role in the complicated CN signalling network in mouse brain. Short-term treatment of SY5Y cells with CsA had no significant effects on RCAN1 expression but significantly affected SOD1 expression. In these in vitro studies, the cells were exposed to CsA for 1 h. This time period is most likely too short to elicit potential changes in RCAN1 protein expression, as the RCAN1 protein levels are determined by many processes, including RCAN1 transcript expression, post-translational control, and protein degradation, among others.
It is well known that the clinical use of CsA is limited by its toxicity in combination with its narrow therapeutic index (33, 34). Neurotoxicity is one of the most serious side effects and is manifested by confusion, cortical blindness, insomnia, anxiety, amnesia, and coma in patients undergoing CsA treatment for organ transplantation. The dose of 20 mg/kg/day CsA in mouse is equal to approximately 155 mg/60 kg/day CsA in humans, which is acceptable for transplant patients, who often receive CsA doses on the order of 200 mg/46 kg/day (35). In summary, the present data provide partial evidence that CsA neurotoxicity might be partially induced by the changes of RCAN1, SOD1, and CN activity despite the low concentration of CsA in brain. Furthermore, insights into mechanisms responsible for CsA-induced changes of endogenous CN inhibitors may be obtained by a careful study of CN/RCAN1 signalling pathway.
Project 81072648 is supported by the National Nature Science Foundation of China and Project supported by the Fundamental Research Funds for the Central Universities.