| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Internal Medicine (D.F.B., E.P.A., M.C., G.R.S., J.B.C., M.J.A.S., L.A.V.), Medical Genetics (M.G.-C., I.L.-C.), and Physiology and Biophysics (M.H.T., A.C.B.), State University of Campinas, 13083-970 Campinas SP, Brazil; and Department of Physiology and Biophysics (J.V.C.F., L.C.M.) and Instituto do Coração (S.C.), University of Sao Paulo, 05508-900 São Paulo SP, Brazil
Address all correspondence and requests for reprints to: Licio A. Velloso, Departamento de Clínica Médica, Faculdade de Ciências Medicas, State University of Campinas, 13083-970 Campinas SP, Brazil. E-mail: lavelloso{at}fcm.unicamp.br.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In classical insulin signaling, PI3-kinase is activated after its recruitment by tyrosine-phosphorylated insulin receptor substrate (IRS)-1/2 (10, 11). Once activated, PI3-kinase catalyzes the incorporation of phosphate at the 3' position in membrane-bound phosphatidylinositol (PI), generating PI3,4-bisphosphate (PI3,4-P2) and PI3,4,5-trisphosphate (PI3,4,5-P3) (12). These phosphorylated PIs serve as docking sites for downstream signaling proteins such as phosphoinositide-dependent kinase and Akt, which forward the insulin signal toward distal effectors (12). The affinity of plecstrin homology domain-containing proteins toward PI3,4,5-P3 is much higher than their affinity toward PI3,4-P2. As a consequence, inositol 5'-phosphatases (which catalyze the conversion of PI3,4,5-P3 to PI3,4-P2) exert the most potent effect in the control of the intracellular signals that are transduced through these intermediates (13, 14).
SH2-containing inositol 5'phosphatase (SHIP)-1 and -2 are the prototypic inositol 5'phosphatases responsible for shutting down the signal transduction through PI3,4,5-P3. SHIP1 is a 145-kDa protein expressed predominantly in cells of the hematopoietic lineage, responding to cytokine and growth factor stimulation and participating in the control of apoptosis, differentiation, migration, and degranulation (13, 15). In addition, the 150-kDa SHIP2 is ubiquitously expressed and responds predominantly to insulin stimuli (13, 15). The lack of SHIP2 expression leads to improved insulin signal transduction in liver and skeletal muscle (16). Moreover, combined SHIP2/Phox2a gene disruption leads to severe neonatal hypoglycemia and perinatal death, whereas heterozygotic disruption of these genes produces a highly insulin-sensitive phenotype (17). Because the main phenotypic consequence of exclusive SHIP2 gene deletion is resistance to dietary obesity (16), we decided to evaluate the participation of SHIP2 in hypothalamic insulin signaling and its effects on the control of food intake. During the optimization of the methodological approach, we were unable to find substantial SHIP2 (and SHIP1) immunoreactivity in hypothalamic protein extracts. However, using SHIP2 and SHIP1 antibodies, we detected an approximately 70-kDa protein that was highly expressed in the hypothalamus and responded to intracerebroventricular (ICV)-injected insulin by undergoing tyrosine phosphorylation.
Here we show that the enzyme phosphoinositide-specific inositol polyphosphate 5-phosphatase IV (5ptase IV) corresponds to the approximately 70-kDa SHIP2 cross-reactive protein. Upon acute hypothalamic treatment with insulin, 5ptase IV undergoes tyrosine phosphorylation, associates with IRS1, IRS2, and PI3-kinase and becomes active to catalyze the conversion of PI3,4,5-P3 to PI3,4-P2. In addition, by inhibiting the expression of 5ptase IV in the hypothalamus of living rats, we demonstrate that this phosphatase plays an important role in the control of the transduction of the insulin signal in this anatomical site.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Phosphorthioate-modified oligonucleotides
Sense and antisense phosphorthioate oligonucleotides specific for 5ptase IV [sense (Sen5ptase), 5'-ATG CTC CAG GGA CAG CT-3' and antisense (AS5ptase), 5'-AGC TGT CCC TGG AGC AT-3'], SHIP2 (sense, 5'-CTG CGG AGG AGC TGC T-3' and antisense, 5'-AGC AGC TCC TCC GCA G-3'), and SHIP1 (sense, 5'-GGA ACA TGG GTA ATG CAC CC-3' and antisense 5'-GGG TGC ATT ACC CAT GTT CC-3') were produced by Invitrogen. All sequences were selected among two unrelated pairs of oligonucleotides on the basis of their ability to inhibit 5ptase IV protein expression, as evaluated by immunoblotting of total protein extracts of hypothalamus using the anti-5ptase IV antibody. The SHIP2 antisense oligonucleotide was thoroughly characterized in a previous study (14). All the antisense oligonucleotide sequences were submitted to nucleotide-nucleotide BLAST analyses (18) (www.ncbi.nlm.nih.gov) and matched only for the Rattus norvegicus 5ptase IV coding sequences.
Production of a polyclonal antibody to 5ptase IV
Five male New Zealand White rabbits were immunized with a peptide specific for 5ptase IV (N-YVLLSSAAHGVLYM-COOH). The sequence of the peptide was submitted to protein-protein BLAST analysis (www.ncbi.nlm.nih.gov) and matched for Rattus norvegicus, Mus musculus, and Homo sapiens 5ptase IV amino acid sequence and with no overlapping with SHIP2 and SHIP1. IgG (100 µg/ml) obtained from the rabbit producing the best response was affinity purified using the Protein A AffinityPak column (Pierce Biotechnology, Inc., Rockford, IL). The specificity of the antibody was evaluated by immunodepletion with the specific peptide.
Experimental animals and research protocols
Eight-week-old male Wistar rats (Rattus norvegicus) from the University of Campinas Central Animal Breeding Center were used in the experiments. The rats were allowed access to standard rodent chow and water ad libitum. Food was withdrawn 12 h before the signal transduction experiments or 6 h before the experiments for determination of food intake. All experiments were conducted in accord with the principles and procedures described by the National Institutes of Health (NIH) Guidelines for the Care and Use of Experimental Animals and were approved by the State University of Campinas Ethical Committee.
During the initial steps of the study, hypothalami were obtained from non-ICV-cannulated rats. Protein extracts from these hypothalami were used in regular immunoblot experiments to evaluate the expression of SHIP2 (and SHIP1). In parallel, ICV-cannulated rats were treated in time-course experiments with 10–6 M insulin and the hypothalami were obtained for immunoprecipitation and immunoblot analysis of insulin receptor (IR), IRS2, and SHIP2 tyrosine phosphorylation. Similar experiments were performed to evaluate the time course of insulin-induced tyrosine phosphorylation of 5ptase IV. For the remainder of the experiments, the ICV-cannulated rats were randomly divided into three groups: 1) control (Ctr) rats treated with saline (2.0 µl, two daily doses for 4 d); 2), 5ptase IV antisense (AS5ptase) rats treated with 5ptase IV antisense oligonucleotide (2.0 µl containing 2.0 nmol antisense oligonucleotide, two daily doses for 4 d); and 3) 5ptase IV sense (Sen5ptase) rats treated with 5ptase IV sense oligonucleotide (2.0 µl containing 2.0 nmol sense oligonucleotide, two daily doses for 4 d). Some rats were treated ICV with SHIP2 and SHIP1 sense or antisense oligonucleotides (2.0 µl containing 2.0 nmol oligonucleotide, two daily doses for 4 d) for evaluation of inositide trisphosphate accumulation in hypothalamus. For evaluation of 5ptase IV oligonucleotide efficiency, some rats were ICV treated with a single dose of AS5ptase and after 0, 1, 3, 6, 12, and 24 h were used for determination of 5ptase IV expression in hypothalamus by regular immunoblot experiments. For evaluation of the SHIP2 and SHIP1 oligonucleotide efficiency, some rats were treated for 4 d with two doses of the sense or antisense oligonucleotides through ip injection. At the end of the experimental period, fragments of skeletal muscle or thymus were obtained and used in regular immunoblot experiments, as described elsewhere (14).
All signal transduction experiments were performed on the morning of the fifth day in 12 h-fasted rats. For evaluation of insulin-induced inhibition of spontaneous food intake, the chow was withdrawn at 12 h (noon) on the fourth day, and the protocol was started at 18 h on the same day [in these experiments some rats from each group were ICV treated with the PI3-kinase inhibitor LY 294002 (2.0 µl of a solution containing 50 µmol/liter) at 1730 h]. Some rats were used for immunohistochemical analysis of 5ptase IV, SHIP1, and SHIP2 expression in hypothalamus. All protocols were performed with groups of four to six rats per condition tested.
ICV cannulation
For the treatment with antisense oligonucleotide, evaluation of insulin-induced inhibition of food intake, and evaluation of insulin-signal transduction, rats were stereotaxically instrumented using a Stoelting stereotaxic apparatus, according to a method previously described (9, 19). Coordinates were: anteroposterior, 0.2 mm/lateral, 1.5 mm/depth, 4.0 mm. Cannula efficiency was tested 1 wk after cannulation by the evaluation of the drinking response elicited by ICV angiotensin II (19). After the experiments, cannula placement was also evaluated by histology.
Protocol for food ingestion determination
ICV-cannulated rats were food deprived for 6 h (from 12 to 18 h) and at 18 h were ICV treated with insulin (2.0 µl, 10–6 M), leptin (2.0 µl, 10–6 M), or saline (2.0 µl). Food ingestion was determined over the next 12 or 24 h.
Metabolic characterization of experimental animals
For evaluation of the effect of the inhibition of hypothalamic 5ptase IV expression on metabolic parameters, groups of five rats were ICV treated with saline (2.0 µl, twice a day) or AS5ptase (2.0 µl, 2.0 nmol, twice a day). After 10 d the animals were anesthetized, blood was collected, and the carcass was used for evaluation of body fat. Plasma glucose was determined by the glucose oxidase method (20), RIA was used to measure serum insulin according to a previous description (21), and leptin concentrations were determined using a commercially available ELISA kit (Crystal Chem Inc, Chicago, IL). Body carcass composition was determined as previously described (22).
PCR cloning and sequencing of 5ptase IV
Five fragments of conserved domains of the Rattus norvegicus 5ptase IV cDNA were obtained by RT-PCR using the following primer sequences: 5'-CAT GGA AAT CAG GGT CTG TG-3', 5'-CCG GGT TTC TAG AGT GAT AC-3', 5'-CCC ATG GTT TTC CAG AGT AC-3', 5'-CTC CTC CGG TTC ATT CAG AT-3', 5'-CCC AGC TCT CAT TTC TCA CA-3' 5'-GAA GCC ATT CAA AGG ACG GA-3', 5'-ATC TGG AGG GAA GTC TTC TG-3', and 5'-TCA AGA CAC AGT GCA CAC AG-3'. All five cDNA fragments were subsequently cloned into a pGEM-T Easy (Promega) vector, sequenced (using SP6 and T7 primers), and sequence homologies evaluated by BLASTn (18).
Tissue extraction, immunoblotting, and immunoprecipitation
ICV-cannulated rats were anesthetized and acutely treated with saline (2.0 µl) or insulin (10–6 M, 2.0 µl). After 2 min (for IR), 5 min (IRS2 and PI3-kinase), and 13 min (or according to the time-course experiments, as presented in Results and figures) (for SHIP2 and 5ptase IV), the hypothalami were obtained and immediately homogenized in solubilization buffer at 4 C [1% Triton X-100, 100 mM Tris-HCl (pH 7.4), 100 mM sodium pyrophosphate, 100 mM sodium fluoride, 10 mM EDTA, 10 mM sodium orthovanadate, 2.0 mM phenylmethylsulfonyl fluoride, and 0.1 mg aprotinin per milliliter] with a Polytron PTA 20S generator (model PT 10/35; Brinkmann Instruments, Westbury, NY). Insoluble material was removed by centrifugation for 20 min at 9000 x g in a 70.Ti rotor (Beckman Instruments, Fullerton, CA) at 4 C. The protein concentration of the supernatants was determined by the Bradford dye binding method. Aliquots of the resulting supernatants containing 2.0 mg of total protein were used for immunoprecipitation with antibodies against IR, IRS2, SHIP2, and 5ptase IV at 4 C overnight, followed by SDS-PAGE; transfer to nitrocellulose membranes; and blotting with antiphosphotyrosine, anti-IR, anti-IRS1, anti-IRS2, anti-p85/PI3 kinase, anti-SHIP2, and anti-5ptase IV. In direct immunoblot experiments, 0.2 mg of protein extracts were separated by SDS-PAGE, transferred to nitrocellulose membranes, and blotted with anti-IR, anti-IRS2, anti-SHIP1, anti-SHIP2, and anti-5ptase IV antibodies, as described (23). In immunodepletion experiments, aliquots of hypothalamus total protein extracts containing 0.5 mg protein were used in three consecutive rounds of immunoprecipitation using either anti-SHIP2 or anti-5ptase IV antibodies. At each immunoprecipitation round, a sample of the supernatant was obtained and used in parallel with a sample of total protein extract in SDS-PAGE and immunoblotting analysis using the anti-5ptase IV antibody.
Real-time quantitative PCR
Six rats were anesthetized, decapitated, and their frontal cortex, hippocampus, hypothalamus, midbrain, striatum, and cerebellum immediately dissected on ice and snap frozen in liquid nitrogen. Frozen samples were immersed in TRIzol and homogenized for 30 sec at maximum speed. Total RNA was isolated according to the manufacturer’s guidelines and quantified by a spectrophotometer. The integrity of RNA was verified on ethidium bromide-stained 1% agarose gel, and the fluorescence intensity ratio of 28S/18S rRNA was determined (Eagle Eye; Stratagene, La Jolla, CA). Only samples that met our criteria of quality (both 260/280 nm and 28S/18S > 1.8) were included in the experiments. RNA samples were DNase treated at 37 C for 30 min and the enzyme inactivated at 65 C for 10 min. All 36 samples were column purified, and 1.6 µg of total RNA were reverse transcribed at the same time using a master mix containing oligo(dT) primer and SuperScript III in a final volume of 20 µl. The following pairs of primers were used to amplify a product of 96 bp corresponding to the 5ptase IV gene of interest (gi 62644778 ): sense 5'-CGA GGT TCT GGG TCT TCT G-3' and antisense 5'-ATT GGG GTG CTG ACT TTG A-3'; a product of 162 bp for the Gapd reference gene (NM_017008) sense 5'-ATG GTG AAG GTC GGT GTG-3' and antisense 5'-GAA CTT GCC GTG GGT AGA G-3'; a product of 172 bp for the Actb reference gene (NM_031144) sense 5'-CGT TGA CAT CCG TAA AGA CC-3' and antisense 5'-GCC ACC AAT CCA CAC AGA-3'; and a product of 101 bp for the Ppia reference gene (NM_017101) sense 5'AAT GCT GGA CCA AAC ACA AA-3' and antisense 5'-CCT TCT TTC ACC TTC CCA AA-3'.
The specificities of the products were confirmed by BLAST analyses and electrophoresis on an ethidium bromide-stained 3% agarose gel. Real-time PCR analysis of gene expression was carried out in an ABI Prism 7700 sequence detection system (Applied Biosystems). The optimal concentration of cDNA and primers as well as the maximum efficiency of amplification were obtained through five-point, 2-fold dilution curve analysis for each gene. Each PCR contained 2.5 or 5.0 ng of reverse-transcribed RNA (depending on the gene), 200 nM of each specific primer, SYBR Green PCR master mix, and RNase free water to a 20 µl final volume. cDNA samples from all brain areas were processed at the same time in triplicate for each gene and the negative controls included for each brain area/primer. The PCR conditions were 10 min at 95 C, followed by 40 cycles at 95 C for 15 sec and 60 C for 60 sec, and a melting step (dissociation curve) was performed after each run to further confirm the specificity of the products and the absence of primer dimers. Real-time data were analyzed using Sequence Detector System 1.7 (Applied Biosystems). The relative expression of 5ptase IV among brain areas was calculated according to a previous study (24).
PI3-kinase activity assay
For IRS2-associated PI3-kinase activity assays, 5.0 mg of total protein extracts obtained from the hypothalami of rats exposed to specific experimental conditions were handled using a previously described method (10, 11). Phosphorylated phosphatidylinositol was separated by TLC and specific dots were visualized by exposing the TLC plates to RX-films. Dot intensities were quantified by digital densitometry of the developed autoradiographs.
PI4,5-P2 5-phosphatase assay
For determination of 5ptase IV catalytic activity toward 5' phosphorylated inositol, we adapted a previously described method (25). In short, 5ptase IV was immunoprecipitated from samples (2.0 mg total protein) of hypothalami of ICV-cannulated rats treated with saline or AS5ptase and with or without insulin, as described above. Protein A-Sepharose pellets were resuspended in 50 µl solution containing 20 mM Tris (pH 7.2), 150 mM NaCl, 200 µg/ml BSA, 3.0 mM MgCl2, 2.0 mM cetyltrimethylammonium bromide, and 250 mM [3H]PI4,5P2 (3500 cpm/nmol). Reactions were performed for 30 min at 37 C and terminated by the addition of 500 µl of 2.0 M KCl and 200 µl chloroform/methanol (1:1). Products were separated by TLC in chloroform/methanol/acetic acid/H2O (43:38:4:7). Lipids in TLC plates were visualized by iodine staining and compared with the migration of known standards PI, PI4-P1, and PI4,5-P2. Products of the reaction were excised from the TLC plate and quantified by liquid scintillation counting. A hypothalamic protein extract sample treated with no 5ptase IV antibody but only with protein A-Sepharose was used as a control. Results are presented as the ratio between the products of PI4,5-P2/PI 4P1 obtained in the analyzed sample and in the control (Nab) sample.
Anion exchange HPLC analysis of PIs
The method used for evaluation of PIs has been previously published (14, 26, 27, 28). In brief, pools of two hypothalami were obtained and homogenized in 1.8 ml of a solution composed of methanol-chloroform-HClO4 (8%) (20:10:1). After the addition of 500 µl chloroform and 500 µl of 1% HClO4, the lower organic phase was collected, washed twice with 1% HClO4, and evaporated. Deacylation was performed as described previously (26). The deacylation product was resolved on an anion-exchange column (Sodex Ashipak, E8502 N7C) with a gradient of 980 µM (A), 3 M NaH2PO4 (B) (pH 3.8). The linear gradient rose to 7% over 30 min, and a 1-min step to 15% followed by a linear gradient to 30% B at 60 min, followed by a linear gradient to 60% B at 80 min. Finally, buffer B was increased to 100% over 5 min. Highly purified nonphosphorylated PI, PI5-P1, PI4,5-P2 and PI3,4,5-P3 were used as standard controls.
Waters HPLC with refractive index detector
Analysis was performed as previously described, with minor modifications (29). Shortly, samples prepared as stated above under Anion exchange HPLC analysis of PIs section, and the standards PI, PI5-P1, PI4,5-P2, and PI3,4,5-P3 were analyzed using a Waters HPLC with refractive index detector composed of 515 HPLC pumps, a Millennium chromatographic management, and refractive index 2414 with a refractive index detector (Waters Corp., Milford, MA). The reverse-phase column was equilibrated with 0.015 M formic acid solution containing 60% MeOH and 0.4% tetrabutylammonium hydroxide (pH 4.3), and the samples (20 µl) were resolved through the column that was heated at 35 C. The refractive indexes were used as arbitrary units for comparative and semiquantitative analysis.
Confocal microscopy
Paraformaldehyde-fixed hypothalami were sectioned (5 µm) and used in regular single- or double-immunofluorescence staining using anti-IR (1:20), anti-SHIP2 (1:50), anti-SHIP1 (1:50), Anti-NeuN (1:100), and anti-5ptase IV (1:50) antibodies, according to a previously described protocol (30). Analysis and photodocumentation of results were performed using a LSM 510 laser confocal microscope (Zeiss, New York, NY). The anatomical correlations were made according to the landmarks given in a stereotaxic atlas (31).
In situ hybridization
For tissue preparation, naïve rats were deeply anesthetized, and immediately after respiratory arrest, thoracotomy was performed to allow transcardiac perfusion (20 ml/min) with 0.01 M PBS (150 ml), followed by 4% paraformaldehyde in 0.1 M PBS (pH 7.2) for 60 min. The brain was removed, stored for 4 h in the same solution of paraformaldehyde, and then immersed in fixative solutions with sucrose gradients (20–30%). Specimens were then stored at –80 C until processing. Serial sections (20 µm) of the hypothalami were obtained according to the landmarks given in a stereotaxic atlas (31). Experiments were conducted in duplicates: for each rat (n = 4), slices of two wells were hybridized with antisense and two with sense (negative controls) 35S-oligonucleotides for 5ptase IV. Hybridization experiments were performed as described previously (32). Briefly, oligodeoxynucleotide probes complementary to the rat nucleotide sequence 2121–2141 [National Center for Biotechnology Information (NCBI)-NIH gi 62644778], or the respective sense sequence, were labeled with 35S dATP using terminal deoxynucleotidyl transferase (Invitrogen). After removing nonincorporated radioactivity, probes were diluted in hybridization buffer to yield approximately 0.3–0.5 x 106 cpm per 100 µl. The hybridization reaction was performed in free-floating sections (
12–16/tube) incubated overnight with 35S-probe (300 µl), followed by high-stringency posthybridization treatment. Slices, mounted on polylysine-covered slides, were submitted to autoradiography (exposition to NTB-2 liquid emulsion (Kodak, Rochester, NY) with development 3 wk later) and counterstained with cresyl violet. Hypothalamic areas were identified by bright field examination. After localizing the target area, dark-field pictures (72336 µm2 window) were obtained at x200 magnification. Density measurements (arbitrary units) were made on the acquired dark-field images (Image Pro Plus; Media Cybernetics, Silver Spring, MD). Background measurements were taken from adjacent areas showing no labeling.
Statistical analysis
Specific protein bands present on the blots or PIP dots on TLC plates were quantified by densitometry. Mean values ± SEM obtained from densitometric scans were compared using Student’s t test for paired samples or by repeat-measures ANOVA (one- or two-way ANOVA) followed by post hoc analysis of significance (Bonferroni test) when appropriate. For real-time PCR, the resulting mean values of 5ptase IV in each brain area were compared using ANOVA followed by Student-Newman-Keuls pos hoc test because they were normally distributed. P < 0.05 was accepted as statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, SHIP2 immunoreactivity was colocalized with the IR in neurons of the arcuate and lateral nuclei of the hypothalamus. A similar pattern of staining was obtained when using a SHIP1-specific antibody (Fig. 1B). However, when hypothalamic protein extracts were used in regular immunoblotting experiments with anti-SHIP2 antibodies (Fig. 1C, left-hand blot), only a faint band was observed at approximately 150 kDa, which was of a much lesser magnitude than the typical SHIP2 immunoreactivity detected in skeletal muscle protein extracts. In addition, the same pattern of a faint approximately 145-kDa band was observed when blotting SDS-PAGE separated hypothalamic protein extracts with SHIP1 antibodies (Fig. 1C, right-hand blot). Once more, the band detected in hypothalamic extracts was much fainter than the typical SHIP1 immunoreactivity detected in thymic protein extracts. Interestingly, both SHIP2 and SHIP1 antibodies detected an approximately 70-kDa protein band in all tissues evaluated (Fig. 1C). To determine whether the SHIP2-immunoreactive approximately 70-kDa protein would respond to insulin by undergoing tyrosine phosphorylation, Wistar rats were submitted to lateral ventricle ICV cannulation and, after a recovery period, were chased with a single dose of insulin ICV. As shown in Fig. 1D, insulin was able to induce the tyrosine phosphorylation of the SHIP2-immunoreactive approximately 70-kDa protein following a time-dependent pattern, peaking at 13 min and returning to basal level at 30 min.
|
. The alignment of all three proteins revealed that similarities occurred in the middle portions of the mature proteins and were overlapping in 5ptase IV vs. SHIP1 and 5ptase IV vs. SHIP2 alignments (Fig. 2A). To confirm whether the high degree of identity between the proteins could explain the cross-reactivity of SHIP1 and SHIP2 antibodies with 5ptase IV, we immunized rabbits with a 5ptase IV-specific peptide and produced a polyclonal antiserum against this phosphatase. As depicted in Fig. 2B, the serum from an immunized rabbit recognized a 72-kDa protein that could be depleted by previous sequential immunoprecipitations with SHIP2 and SHIP1 antibodies as well as the 5ptase IV antibody (Fig. 2C). Different from the SHIP2 antibody that cross-reacts with 5ptase IV in skeletal muscle (Fig. 2D, lane1) and hypothalamus (Fig. 2D, lane 5) and from the SHIP1 antibody that cross-reacts with 5ptase IV in thymus (Fig. 2D, lane 2) and hypothalamus (Fig. 2D, lane 6), the 5ptase IV antibody does not recognize bands at approximately 145–150 kDa in skeletal muscle (Fig. 2D, lane 3), thymus (Fig. 2D, lane 4), and hypothalamus (Fig. 2D, lane 7), suggesting that it lacks cross-reactivity with SHIP2 and SHIP1. The cross-reactivity between the antibodies was further evaluated by double-staining immunohistochemistry and confocal microscopy. Using this method the immunoreactivity for SHIP2 and 5ptase IV and for SHIP1 and 5ptase IV was detected in overlapping cells in the arcuate nucleus (Fig. 2E). Next, using the 5ptase IV antiserum, we were able to detect the presence of this enzyme in neurons of the arcuate nucleus (Fig. 3A) and lateral hypothalamus (Fig. 3C). Particularly in the arcuate nucleus, but also to a minor degree in the lateral hypothalamus (not shown), 5ptase IV immunoreactivity coincided with the expression of the IR (Fig. 3B). The 5ptase IV immunoreactivity was also detected in neurons of the frontal cortex (Fig. 3D, upper lane), parietal cortex (Fig. 3D, middle lane), and Purkinje neurons of the cerebellum (Fig. 3D, lower lane).
|
|
, E and F, respectively). In addition, real-time PCR demonstrated the expression of the 5ptase IV transcript in the frontal cortex, striatum, hippocampus, hypothalamus, midbrain, and cerebellum (Fig. 3G). The highest expressions were seen in the cerebellum and hypothalamus. Finally, to confirm the expression of 5ptase IV in hypothalamus, we performed PCR cloning, followed by sequencing of a substantial fragment of the mRNA coding for this protein. As depicted in Table 1, the BLASTn alignment of the cDNA fragment obtained by RT-PCR confirms the identity of the protein, which displayed 97% homology with the Rattus norvegicus 5ptase IV cDNA sequence, gi 62644778, available at the GenBank.
|
, insulin was capable of inducing the tyrosine phosphorylation of the IR (Fig. 4A, left-hand upper blot, 2 min) and IRS2 (Fig. 4A, right-hand upper blot, 5 min). ICV insulin was also capable of promoting the association of IRS2 with the p85 subunit of PI3-kinase (Fig. 4A, left-hand lower blot, 5 min) and the catalytic activation of PI3-kinase (Fig. 4A, right-hand lower image of TLC, 5 min). In addition, ICV insulin was capable of inducing, in a time-dependent fashion, the tyrosine phosphorylation of 5ptase IV without affecting its protein amount (Fig. 4B). Similar to the results obtained using the SHIP2 antibody (as in Fig. 1D), insulin-induced tyrosine phosphorylation of 5ptase IV peaked at 13 min and returned to basal levels at 30 min. To determine whether the insulin treatment would induce any physical interaction between 5ptase IV and proteins that belong to the insulin signaling pathway, samples of hypothalamic protein extracts were obtained from ICV-cannulated rats treated with insulin according to the same time course as used for evaluation of insulin-induced tyrosine phosphorylation and used in immunoprecipitation experiments. As depicted in Fig. 4C, insulin treatment led to the molecular interaction of 5ptase IV with the p85 subunit of PI3-kinase (Fig. 4C, upper blot), IRS2 (Fig. 4C, second blot), and IRS1 (Fig. 4C, third blot) but not Akt (Fig. 4C, fourth blot).
|
, the ICV treatment of rats with a single dose of the 5ptase IV antisense oligonucleotide (AS5ptase) promoted a progressive reduction of 5ptase IV expression in hypothalamus, which was apparent at 6.0 h and lasted for at least 24 h. As shown in Fig. 4E, the ICV treatment of rats with AS5ptase but not the 5ptase IV sense oligonucleotide (Sen5ptase) was capable of significantly reducing the protein expression of 5ptase IV (79 ± 13% reduction, P < 0.05) without affecting the expression of other proteins that participate in the transduction of the insulin signal, such as IR, IRS2, and p85-PI3-kinase. The inhibition of 5ptase IV expression could also be observed by immunohistochemistry because no staining of neurons of the arcuate (Fig. 4F) and lateral (not shown) nuclei of the hypothalamus was obtained with the 5ptase IV antibody. The effect of the AS5ptase injected through an ICV cannula placed in the lateral ventricle was confined to the hypothalamus, promoting no significant inhibition of 5ptase IV expression in the frontal and parietal cortexes as well as in the cerebellum (Fig. 4G). Inhibition of 5ptase IV expression significantly reduced the insulin-induced tyrosine phosphorylation of 5ptase IV (Fig. 5A), and interestingly, it also significantly increased the basal IRS2-associated catalytic activity of PI3-kinase without affecting its IRS2-associated, insulin-induced activity toward phosphatidylinositol (Fig. 5B). In addition, we observed that the catalytic activity of 5ptase IV toward PI4,5-P2 was stimulated, following a similar time course to its insulin-induced tyrosine phosphorylation (Fig. 5C), suggesting that the tyrosine phosphorylation event plays a role in the control of the functional activity of the enzyme.
|
). Next, the evaluation of the effect of 5ptase IV expression inhibition on the levels of inositide trisphosphate in hypothalamic extracts was performed by anion exchange HPLC and Waters HPLC with a refractive index detector. As depicted in Fig. 6A, both methods were capable of detecting the standards for PI5-P1, PI4,5-P2, and PI3,4,5-P3. The treatment with AS5ptase significantly increased the amount of inositide trisphosphate in hypothalamic extracts, a phenomenon that is not reproduced by inhibition of SHIP2 or SHIP1 expression by antisense oligonucleotides (Fig. 6B). In addition, after insulin treatment, there is an increase in inositide trisphosphate, which is detectable as a peak at approximately 40 min. In hypothalamic extracts from rats treated with AS5ptase, the inositide trisphosphate peak is already present in basal conditions (saline treated), and no further increase is detected after insulin treatment (Fig. 6C).
|
shows that the inhibition of 5ptase IV significantly reduced basal 12 h-spontaneous food intake. However, this treatment did not modify 12-h food intake after ICV insulin stimulus, compared with the control. In all experimental groups, including AS5ptase-treated rats, the use of LY 294002 completely inhibited insulin- or AS5ptase-induced inhibition of food intake, suggesting that most of the effect herein described depends on the formation of PI3,4,5-P3 catalyzed by PI3-kinase. The inhibition of hypothalamic 5ptase IV expression was also efficient to reduce 24-h spontaneous food intake without affecting the effects of insulin and leptin on this parameter (Fig. 7B). In addition, the treatment of rats with AS5ptase for 10 d promoted a dramatic fall in mean daily food intake (Fig. 7C) and by a significant and progressive reduction in body weight (Fig. 7C), which was accompanied by a significant reduction in relative body fat (Table 2). No changes on blood glucose, insulin, and leptin levels were promoted by the inhibition of 5ptase IV for 10 d (Table 2). Also, we observed no modification in the general behavior of the animals treated with the AS5ptase. The reduction of feeding was not due to an apparent aversive effect and was not accompanied by modification in the macroscopic aspect of the feces. Finally, the inhibitory effect of AS5ptase treatment on mean daily food intake was reversible because it was completely abolished 3 d after interruption of the treatment.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Of special interest, we observed that in all immunoblot experiments using either SHIP2 or SHIP1 antibodies, a band of approximately 70 kDa could be detected. This protein was later shown to respond to ICV stimulus with insulin by undergoing time-dependent tyrosine phosphorylation. Using the NCBI-NIH engine for the search of similarities between deposited protein sequences, we found the 72-kDa 5ptase IV protein, which shares 32 and 30% similarity with SHIP2 and SHIP1, respectively. A polyclonal antibody reacting against the 72-kDa 5ptase IV was produced, and the labeling of neurons of the arcuate and lateral hypothalamic nuclei with this antibody, accompanied by the cross-reactivity of 5ptase IV with SHIP2 (and SHIP1), as evidenced by immunodepletion experiments, plus the sequencing of a considerable fragment of 5ptase IV from hypothalamic mRNA and, finally, the detection of 5ptase IV in hypothalamus by in situ hybridization and real-time PCR, strongly suggests that this phosphatase corresponds to the SHIP2 approximately 70-kDa cross-reactive protein in the hypothalamus.
A thorough search for gene and overlapping sequences was performed and revealed that SHIP1, SHIP2, and 5ptase IV are encoded by three distinct genes. In Rattus norvegicus, SHIP1 (previously referred simply as SHIP) is encoded by a gene localized at 9q35 (gi 1777941 at the GenBank) that produces a 145-kDa protein expressed mainly in lymphatic and hematopoietic tissues. SHIP2 is encoded by a gene localized at 1q32 (gi 5263146 at the GenBank) and renders a product of 150 kDa expressed in most mammalian tissues. Finally, 5ptase IV, also known as INPP5e, is encoded by a gene localized at 3p13 (gi 62644778 at the GenBank). This gene may produce a large protein of approximately 310 kDa and a smaller form of 72 kDa. Kisseleva et al. (37, 38) and Kong et al. (39) characterized several aspects of the functional activity of the 72-kDa form of 5ptase IV. To our knowledge, no characterization of the expression and activity of the 310-kDa form has been published. We performed longer SDS-PAGE runs to search for the longer form in rat hypothalamic extracts, but no specific band could be detected at this molecular weight (not shown). Thus, we believe that only the 72-kDa form is present in the central nervous system. In fact, in the study by Kong et al. (39), high expression of the 72-kDa form of this enzyme was detected in brain and testis. In the brain, the expression was localized to regions very similar to the ones we report here. Yet according to Kisseleva et al. (37, 38), the mRNA coding for 5ptase IV is detected in tissues such as brain, heart, testis, kidney, pancreas, and liver, appearing as three distinct splicing variants of 9.5, 4.9, and the predominant 3.6 kb. Conversely, Kong et al. (39) reported a mRNA form of 3.9 kb expressed predominantly in testis but to a minor degree in a number of other tissues. These mRNAs code for a mature protein of 644 amino acids (Mr = 70,023), which is resolved as a 72-kDa band in SDS-PAGE. At the functional level, 5ptase IV has the highest affinity for PI3,4,5-P3 (Michaelis constant = 0.65 mM) being up to10-fold greater than the other known phosphatases targeting this substrate (37, 38).
The phosphatidylinositol signaling system participates in the control of a number of important functions in eukaryotes, ranging from the modulation of intracellular calcium levels to the control of glucose uptake and apoptosis (40). The termination of the signal transduction through this system depends, largely, on the activity of specialized inositol phosphatases that are grouped according to their substrate specificity (15). Inositol polyphosphate 5-phosphatases target water-soluble and membrane-anchored inositol phosphates and are subclassified as groups I-IV (for an overview of the system, see Ref. 15). SHIP1 and SHIP2 belong to the group III inositol 5-phosphatases as they hydrolyze water-soluble I1,3,4,5-P4 and lipidic PI3,4,5-P3, whereas 5ptase IV is, for the time being, the sole component of the group IV inositol 5-phosphatase, characterized by its property of targeting PI3,4,5-P3 and forming a complex with PI3-kinase (15).
To pursue our initial objective, we decided to evaluate whether 5ptase IV would participate somehow in the regulation of PI3-kinase-dependent signaling in hypothalamus. First, we confirmed that this enzyme responds to insulin stimulus by undergoing tyrosine phosphorylation within a time frame that follows classic insulin signal transduction through IRS2 and PI3-kinase (41). Insulin is known to induce the tyrosine phosphorylation of SHIP2 in other tissues, but the role of this posttranslational modification in SHIP2 function is not completely elucidated (14, 42, 43, 44). Next, we observed that, in parallel to the phenomenon of insulin-induced tyrosine phosphorylation, 5ptase IV transiently associates with p85 PI3-kinase, IRS2, and to a minor degree with IRS1. In addition, after a similar time course, the catalytic activity of the enzyme toward 5' phosphorylated phosphatidylinositols is stimulated.
Next, we used an antisense oligonucleotide to inhibit 5ptase IV expression exclusively in hypothalamus. The treatment with the oligonucleotide significantly reduced the expression of the target protein without affecting the expression of other proteins of the insulin signaling pathway. Reduction of 5ptase IV expression was accompanied by a reduced insulin-dependent tyrosine phosphorylation of this protein and an increase in basal activity but not insulin-induced IRS2-associated PI3-kinase activity. Moreover, the inhibition of 5ptase IV expression resulted in increased basal, but not insulin-induced, accumulation of inositide trisphosphate in hypothalamus. Because inositol phosphatases control the accumulation of a product that is downstream from PI3-kinase, it is not clear why the inhibition of such an enzyme should result in an increased activity of the kinase. One possible explanation is that the binding of 5ptase IV to PI3-kinase, as demonstrated by Majerus et al. (15), which was confirmed in our studies, could provide an intrinsic mechanism for the control of signal transduction through this system.
In the final part of the study, we evaluated the role of 5ptase IV in two important physiological phenomena controlled by PI3-kinase signaling in hypothalamus, i.e. food intake and variation of body weight (4, 7, 9). Inhibition of the phosphatase expression reduced spontaneous but not insulin- or leptin-inhibited 12- and 24-h food intake. The acute ICV injection of insulin or leptin is known to promote 40–55% reduction in 12- and 24-h food intake (8, 9, 33). This event is largely dependent on PI3-kinase because the inhibition of this system has a profound effect on the anorexigenic properties of insulin and leptin (7, 9). The fact that the inhibition of 5ptase IV expression affected only basal and not insulin- or leptin-inhibited food intake seems to be in keeping with the effect of this treatment only on basal (and not hormone stimulated) PI3-kinase activity and basal accumulation of inositide trisphosphate in hypothalamus. Thus, it seems logical to suggest that the amount of inositide trisphosphate in the hypothalamus plays a direct role in the control of food intake and that the means used to inhibit inositide trisphosphate depletion are sufficient to match the total and apparently maximal effect obtained by insulin (or leptin) treatment to promote the production of this signaling intermediary. This hypothesis is further confirmed by the effect of a longer (10 d) period of ICV treatment in rats with the 5ptase IV antisense oligonucleotide. This treatment resulted in a marked reduction in daily food intake and a parallel reduction in body weight and body fat.
Although we have explored mostly the effects of insulin upon this signaling system, we acknowledge that leptin is also known to use the PI3-kinase signaling pathway to control food intake. However, in a series of recent studies that reinforced the role of the hypothalamic IRS (particularly IRS2)/PI3-kinase system in the control of food intake and body weight (6, 8, 45, 46, 47), it was shown that, similar to the Janus kinase 2/signal transducer and activator of transcription 3 signaling pathway, PI3-kinase is a point of convergence between the insulin and leptin signaling systems in the hypothalamus. The effects of these hormones on both these pathways are rather specific. Leptin exerts its effects predominantly through the Janus kinase 2/signal transducer and activator of transcription 3 system and the effect of insulin on this pathway modulates leptin activity (8, 9). Conversely, hypothalamic signaling through PI3-kinase is predominantly controlled by insulin, with leptin exerting a modulatory role on this pathway (6, 45, 46, 47). Thus, we believe that 5ptase IV may act on signals delivered by both insulin and leptin. However, its importance predominates in the former rather than the later. Further studies will be required to confirm this idea.
In conclusion, we have identified an inositide phosphatase that plays an important role in inositide trisphosphate accumulation in the hypothalamus. A scheme presented in Fig. 8 shows our hypothesis regarding the participation of 5ptase IV in the transduction of the insulin signal in hypothalamus. This phosphatase seems to play, in the hypothalamus, a similar role to that played by SHIP2 in peripheral tissues, i.e. as an off-switch for the insulin signal transduction through PI3-kinase. Moreover, our results confirm the absent or minimal expression of SHIP2 in hypothalamus and provide further evidence (in addition to SHIP1 and SHIP2) for the existence of tissue-specific systems for the control of inositide trisphosphate signaling. Due to the remarkable effect of 5ptase IV on spontaneous food intake and body weight, this molecule may become an interesting target for drug therapy for obesity and related diseases.
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
All the authors of this work have nothing to declare.
First Published Online August 17, 2006
1 D.F.B. and E.P.A. contributed equally to this study. 
Abbreviations: ICV, Intracerebroventricular; IR, insulin receptor; IRS, insulin receptor substrate; PI, phosphatidylinositol; PI3-kinase, phosphatidylinositol 3-kinase; PI3,4-P2, PI3,4-bisphosphate; PI3,4,5-P3, PI3,4,5-trisphosphate; 5ptase IV, inositol polyphosphate 5-phosphatase IV; SHIP, SH2-containing inositol 5'phosphatase; TLC, thin-layer chromatography.
Received March 3, 2006.
Accepted for publication August 4, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, P < 0.05 vs. C (0).
