Alteration in Phosphorylation of P20 Is Associated With Insulin Resistance

Male Wistar rats were fed standard rat diet (NRM Diet 86, Tegel, Auckland, New Zealand) with water ad libitum. [³²P]orthophosphate and d-[U-¹⁴C]glucose were purchased from ICN. 2-deoxy-d-[³H]glucose (1 mCi/ml) was obtained from DuPont-NEN and ¹²⁵I was obtained from Amersham Pharmacia. Human insulin (Actrapid) was obtained from Novo Nordisk. Rat amylin and CGRP were purchased from Bachem (Torrance, CA); epinephrine was from David Bull Laboratories; and dexamethasone was from Sigma. The two-dimensional gel electrophoresis (2-DE) system and reagents were obtained from Pharmacia. Anti-P20 polyclonal antibody was a generous gift from Dr. Kanefusa Kato (25). Anti-GLUT4 (H-61) was obtained from Santa Cruz Biotechnology. The enhanced chemiluminescence (ECL) detection system was from Roche Molecular Biochemicals. The total cellular RNA extraction reagent (TRIZOL), G418, Lipofectamine Plus reagent, and random priming labeling kits were from Life Technology. pCXN2-GLUT4myc, which expresses myc-tagged GLUT4 (GLUT4myc) in mammalian cells, was kindly provided by Dr. David James (University of Queensland, Australia).

All experimental protocols were approved by the Institutional Animal Ethics Committee. Male Wistar rats were injected with dexamethasone (3.1 mg · kg⁻¹ · day⁻¹, i.p.) for 7 days. The weight of rats in both control and dexamethasone-treated groups were monitored daily. By the end of the treatment period, the mean weight of the control group had increased by 12 ± 1%, whereas that of the glucocorticoid-treated group had sharply decreased, by 16 ± 2% (n = 3 experiments, each with three rats per group). Rats were fasted for 18 h before each experiment and were killed by cervical dislocation. Blood was obtained by cardiac puncture from anesthetized animals. The mean blood glucose concentration was 5.4 ± 0.2 and 10.8 ± 0.6 mmol/l in control and dexamethasone-treated rats, respectively, as measured with a YSI 2300STAT glucose/lactate analyzer (Yellow Springs Instruments). The insulin-resistant state of the skeletal muscle was further confirmed using an in vitro d-[U-¹⁴C]glucose incorporation assay, which showed a >95% reduction in the rate of insulin-stimulated glycogen synthesis in dexamethasone-treated rats (results not shown). Insulin and amylin concentrations in the blood of normal and insulin-resistant rats were determined as previously described (17). Rats with insulin resistance induced by chronic high-fat feeding were generated as described previously (26).

Rat extensor digitorum longus (EDL) muscle strips were prepared from 18 h−fasted rats. The dissection and isolation of muscles were carried out under anesthesia with pentobarbital (5–7 mg/100 g body wt, i.p.), as described previously (23). Each muscle was split into three ∼1 mm–wide strips. Muscle strips were preincubated in a shaking coater bath at 30°C for 1 h in 5 ml of Dulbecco’s modified Eagle’s medium (DMEM) without sodium phosphate. All incubation media were gassed with a mixture of 95% O₂ and 5% CO_2. The muscle strips were subsequently transferred to similar flasks containing identical medium plus 0.25 mCi/ml [³²P]orthophosphate and incubated for an additional 4 h to equilibrate the internal ATP pool (23,24). Human insulin, rat amylin, epinephrine, or CGRP were then added to the incubation media for 30 min at stated final concentrations. Reactions were terminated by freezing muscle strips in liquid nitrogen immediately after incubation. Muscle strips were then weighed and stored at −80°C until further analysis.

Muscle strips were homogenized in 2-DE lysis buffer (9 mol/l urea, 2% vol/vol triton X-100, 2% vol/vol pharmalyte [pH 3–10], 200 mmol/l dithiothreitol, 8 mmol/l phenylmethylsulfonyl fluoride) for 5 min on ice. The lysates were briefly sonicated and microcentrifuged at 16,000g for 10 min to remove debris. Protein concentrations of supernatant were determined by the Bradford method, and radioactivity was measured by liquid scintillation counting. ³²P-labeled lysates with equivalent amounts of radioactivity were isoelectrically focused on IPG Drystrip (pH 4–7 and pH 3–10) linear gels using a Multiphor RII electrophoresis system according to the manufacturer’s instructions. Second dimensional SDS-PAGE was carried out using ExcelGel precast 12–14% acrylamide gradient gels. After electrophoresis, the gels were fixed in 10% glacial acetic acid and 40% ethanol, and the proteins were visualized by phosphorimaging or autoradiography. In all figures, the gels are displayed with the acidic end of the isoelectric focusing dimension to the right; the SDS-PAGE direction is from top to bottom.

A full-length cDNA encoding wild-type rat P20 was cloned by reverse transcriptase−polymerase chain reaction using the forward primer 5′GCCCGCGGATCCATGGAGATCCGGGTGCCTGTG3′ and the reverse primer 5′GCCCGGGATCCCTACTTGGCAGCAGGTGGTGAC3′. The resulting clone was validated by DNA sequencing and then inserted into the multiple cloning site of cytomegalovirus promoter-driven eukaryotic expression vector pcDNA3.1 (referred to as pcDNA.P20).

L6 myoblast cells were transfected with pCXN2-GLUT4myc (27) or cotransfected with pCXN2-GLUT4myc and pcDNA.P20 using Lipofectamine Plus reagent (Life Technology) according to the manufacturer’s instructions. Stable transfectants were selected in medium containing the neomycin analogue G418 at 400 μg/ml. At 10 days after transfection, the clones were selected using sterilized steel rings and expanded separately in the presence of G418. Clones that expressed P20 and GLUT4myc were chosen by Western blotting and used for further experiments.

Proteins (∼50 μg) from liver, heart, epididymal fat pad, aortic smooth muscle, EDL muscle, soleus muscle tissue, and whole blood obtained from 18 h−fasted male Wistar rats were separated by SDS-PAGE and subsequently transferred to nitrocellulose membranes. The membranes were blocked overnight at 4°C and then incubated with rabbit anti-P20 polyclonal antibody (1:1,000) for 2 h at room temperature. After incubation with streptavidin-biotinylated horseradish-peroxidase−conjugated secondary antibody for another hour at room temperature, the proteins immunoreactive to the primary antibody were visualized by ECL detection according to the manufacturer’s instructions.

Total cellular RNA was isolated from EDL muscle of 18 h−fasted control and dexamethasone-treated rats using TRIZOL reagent (Life Technology). Then, 15 μg of RNA from each sample were separated by 1.5% agarose-formaldehyde gel electrophoresis and subsequently transferred to Hybond-N⁺ nylon membranes (Amersham Pharmacia Biotech, Uppsala, Sweden) by capillary blotting in 20× sodium chloride−sodium citrate. The P20 cDNA probe was labeled with ³²P-dCTP using a random primer labeling system. The membranes were preincubated with hybridization buffer (0.5 mol/l Na₂HPO₄ [pH 7.2], 10 mmol/l EDTA, 7% SDS) for 3 h at 65°C and subsequently incubated with fresh buffer containing the labeled probe for 18 h. Membranes were then washed, analyzed using a phosphorimager, and quantitated by MacBAS v2.5 software (Fuji Machine Manufacturing, Chiryu, Aichi, Japan). For comparison, RNA samples from EDL muscle strips treated with or without 50 nmol/l amylin were also analyzed in parallel.

L6 cells stably overexpressing GLUT4myc or GLUT4myc plus P20 were grown in six-well plates and differentiated into myotubes in DMEM containing 2% fetal bovine serum for 7 days. The cells were deprived of serum for 16 h before experiments. For glucose uptake assays, L6 myotubes were rinsed three times with Krebs-Henseleit buffer (KHB) and incubated in KHB with or without hormones (insulin or insulin plus amylin) at the indicated concentrations for 15 min at 37°C. Carrier-mediated glucose uptake of 10 μmol/l 2-deoxy-d-[³H]glucose in the above solution was measured for 15 min at 37°C. This was followed by rinsing the cells three times with ice-cold phosphate-buffered saline and cell disruption with 0.1 N NaOH. The associated radioactivity was determined by liquid scintillation counting. The protein concentration was measured with a bicinchoninic acid protein quantitation kit (Pierce). The nonspecific uptake was determined in the presence of 10 μmol/l cytochalasin B and subtracted from each value.

Autoradiography films were scanned and digitized using a Sharp JX-325 scanner, and protein spots were detected, quantitated, and analyzed using the Melanie II software package, version 2.2 (Bio-Rad). The detection parameters were as follows: smooth, 2; Laplacian threshold, 3; partials threshold, 1; saturation, 90; peakedness increase, 100; and minimum perimeter, 10. The matching of multiple features to one feature was not allowed. The pixel value was the optical density (OD). Features were calculated as a percent of the sum of the feature’s volume ([VOL], i.e., the integration of OD over the feature’s area) for all features on the gel. The radioactivity of protein spots was also detected by a phosphorimager and analyzed by MacBAS v2.5 software. The radiation dosage of each spot was displayed in terms of units of photostimulated luminescence (PSL). All of the results presented are based on at least three independent experiments. Statistical analysis was performed using the t test (paired, two-way).

P20 was initially isolated from rat skeletal muscle as a byproduct during the purification of the small heat shock proteins (HSPs) HSP27/28 and αB-crystallin (25). Under normal physiological conditions, it exists as large aggregates. P20 has been thought to be a heat shock−related protein, given that it has significant amino acid sequence similarity with αB-crystallin (47%) and HSP27/28 (35%) (25,28). However, unlike other small HSPs, heat treatment or chemical stress does not induce the expression of P20. Several recent studies have suggested that P20 may be an actin-binding protein that is involved in cyclic nucleotide-mediated vasodilation and relaxation of rat smooth muscle or histamine- and phorbol ester−induced contraction of bovine carotid artery smooth muscle (29,30,31). Interestingly, this protein is also present at high concentrations in circulating whole blood in patients with vascular diseases. It can strongly suppress platelet aggregation in vitro and ex vivo, possibly by inhibiting receptor-mediated calcium influx in platelets (32). However, the precise physiological functions of P20 are still uncertain.

Analysis of the protein content of P20 by Western blot showed that this protein is mainly expressed in rat soleus muscle, EDL muscle, and heart muscle tissues, which account for 35.1 ± 3.2, 29.6 ± 2.7, and 23.3 ± 2.5% of the total P20 in all the tested tissues, respectively (n = 3; means ± SD) (Fig. 1). A small amount of this protein was also detected in smooth muscle (4.9 ± 0.6%), adipose tissue (1.9 ± 0.3%), and blood (5.2 ± 0.6%). 2-DE analysis of ³²P-labeled rat EDL muscle revealed ∼150 phosphoproteins after insulin stimulation (Fig. 2). Quantitative analysis by Melanie II software (Bio-Rad) revealed that P20 is the second most abundant phosphoprotein in insulin-stimulated rat EDL muscle, representing >2% of the total VOL for all features detected. Moreover, P20 is the only detected phosphoprotein that is responsive to both insulin and its antagonists, as analyzed by the proteome approach.

Our previous studies demonstrated that insulin and its antagonists, epinephrine, amylin, and CGRP, elicit differential phosphorylation on different sites of P20, thus producing three phosphorylated isoelectric variants of P20 (termed S1, with an isoelectric point [pI] value of 6.0; S2, with a pI value of 5.9; and S3, with a pI value of 5.6) (23,24). Phosphorylation of S1 occurs at serine 157 of P20, and insulin can increase its phosphorylation through a PI-3 kinase−mediated pathway. Amylin, CGRP, and epinephrine evoke phosphorylation at Ser16 of P20 through a cAMP-mediated pathway, leading to the production of the phosphoisoform S2. In addition, these catabolic hormones also induce the phosphorylation of P20 at another two unidentified sites to produce the phosphoisoform S3.

Here, we further investigated the effect of the interplay between insulin and several of its antagonists on the phosphorylation of P20. Interestingly, we found that insulin and amylin can antagonize each other’s actions in the phosphorylation of this protein (Fig. 3). On the one hand, insulin-induced phosphorylation of S1 was significantly decreased in the presence of amylin. Phosphorylation of S1 in samples treated with 50 nmol/l insulin + 50 nmol/l amylin was 49% lower than that in samples stimulated with 50 nmol/l insulin alone. On the other hand, insulin blocked amylin-evoked phosphorylation of S2 and S3. In the presence of insulin, phosphorylation of S2 and S3 was decreased by ∼72 and ∼74%, respectively, relative to that in muscles treated with amylin alone. However, insulin had no effect on the phosphorylation of S2 and S3 induced by the other two catabolic hormones, epinephrine and CGRP, and vice versa. This result indicated that cross-talk occurs between only the insulin- and amylin-evoked signaling pathways, although all three catabolic hormones are thought to act through G-protein−coupled receptors and to have similar metabolic effects. Amylin inhibited the insulin-evoked PI-3 kinase cascade−mediated phosphorylation of S1. Conversely, insulin suppressed the amylin-evoked cAMP pathway−mediated phosphorylation of S2 and S3. Such an inhibitory effect of insulin on amylin’s biological actions could provide a reasonable explanation as to why administration of exogenous amylin in physiological quantities did not induce hyperglycemia and insulin resistance in some experimental systems.

The fact that insulin has separate effects on the inhibition of biological actions of amylin and CGRP further excludes the possibility that amylin acts solely through a CGRP receptor, although the two peptide hormones are members of the calcitonin-related polypeptide family (33). The amylin-specific receptor still remains to be identified. However, several recent studies have suggested that the identity of an amylin-selective receptor may be determined in part by receptor activity−modifying proteins (34).

We next investigated the phosphorylation patterns of P20 and the effect of insulin and amylin on this protein in dexamethasone-induced diabetic rats with insulin resistance. The diabetic state of these rats was confirmed by the demonstrated loss of body weight, hyperglycemia, and decrease in insulin-stimulated incorporation of glucose into glycogen (results not shown). In dexamethasone-treated rats, the fasted basal plasma concentrations of both insulin (789 ± 94 vs. 203 ± 28 pmol/l in control rats) and amylin (144 ± 17 vs. 22.7 ± 5.9 pmol/l in control rats) were significantly increased (P < 0.01 in each case).

EDL muscle strips from these rats were radiolabeled with ³²P and treated with or without insulin and amylin. Then, phosphorylation of P20 was analyzed by 2-DE and phosphorimaging (Fig. 4). Under the incubation conditions without hormone stimulation, phosphorylation of S2 and S3 was hardly detected in the nondiabetic control rats (Fig. 4A). By contrast, these two phosphoisoforms were clearly visualized in muscle samples from the insulin-resistant rats (Fig. 4B). Quantitative analysis by phosphorimager and MacBAS software showed that the signals associated with both S2 and S3 in dexamethasone-treated rats were about fivefold higher (Table 1). This phenomenon was also observed in a high-fat−induced insulin-resistant rat model (Fig. 5), suggesting that the increased phosphorylation of two isoforms of P20 (S2 and S3) may be associated with insulin-resistant states in general. Analysis of P20 expression revealed that the mRNA level and protein abundance of P20 was not changed in either the diabetic rats or the amylin-treated muscle strips (Fig. 6). These results indicated that the increased phosphorylation of S2 and S3 was not attributable to the increased expression of P20, but rather to a possible defect in the intracellular signal transduction pathways that lead to generation of its phosphorylated isoforms.

Another major alteration in insulin-resistant rats is insulin’s ability to inhibit amylin-evoked phosphorylation of S2 and S3. In normal rats, 50 nmol/l insulin decreased phosphorylation of S2 and S3 by 71.6 and 73%, respectively, compared with that in samples treated with 50 nmol/l amylin alone (Figs. 4E and G). In diabetic rats, on the other hand, amylin-evoked phosphorylation of S2 and S3 was little affected by insulin (Figs. 4F and H). Under this condition, the radioactivity of both S2 and S3 was ∼3.3-fold higher than that of nondiabetic control rats (Table 1).

Insulin resistance is a well-known effect of glucocorticoid excess, but its mechanisms are still uncertain (35). Although muscle is quantitatively the most important tissue for glucose disposal in response to insulin, there are few studies on the effects of glucocorticoids in this tissue. Administration of dexamethasone did not affect the number or affinity of insulin receptors in skeletal muscle but did reduce insulin receptor tyrosine autophosphorylation and decrease IRS-1 activation of PI-3 kinase, suggesting the existence of postreceptor defects (36). It has recently been reported that dexamethasone treatment significantly inhibited the insulin-stimulated translocation of GLUT4 from an intracellular pool to the plasma membrane, although expression of this transporter was paradoxically slightly increased (37).

Pieber et al. (38) observed that whenever diabetes occurred in dexamethasone-treated rats, the level of amylin and the ratio of amylin to insulin were significantly increased. The increase in the amylin-to-insulin ratio was associated with elevated content of proamylin mRNA relative to proinsulin mRNA. Those study results implied that amylin could also be an important contributing factor to the development of dexamethasone-induced insulin resistance. The results of our present study support such a role for amylin. The phosphoisoforms S2 and S3, which were hardly detected in healthy rats but could be induced by amylin, were clearly present in diabetic rats (Fig. 4B). This may have been because of the increased amylin level or the increased amylin-to-insulin ratio. It is interesting to note that in normal rats, insulin specifically suppresses amylin’s actions on the phosphorylation of P20 and elevation of cAMP levels, but has no detectable effect on the actions of two other catabolic hormones, epinephrine and CGRP (Fig. 3). Such an action of insulin was significantly attenuated in dexamethasone-induced diabetic rats (Figs. 4F and H). Based on these results, it is tempting to speculate that under physiological conditions, amylin’s antagonism of insulin-stimulated glucose disposal is inhibited by insulin itself. The impairment of this action by insulin may lead to the enhanced catabolic action of amylin and thus partly cause insulin resistance in dexamethasone-induced diabetic rats.

Although the physiological role of P20 is uncertain, the high abundance of this protein and its diverse responsiveness to insulin and its antagonists suggest that it could be a mediator involved in the biological actions of these metabolic hormones. Notably, P20 has recently been shown to be an actin-binding protein (31). Both cytoskeletal actin filaments and actin-binding proteins have been suggested to play a role in directing traffic of glucose transporters to the cell membrane (39,40). Interestingly, two other proteins whose increased expression may contribute to insulin resistance in type 2 diabetes, Rad and PED/PEA-15, are also cytoskeleton-associated proteins involved in the regulation of glucose transport (41,42). Thus, it is intriguing to speculate that metabolic hormones, such as insulin and amylin, could regulate glucose transport by modulating the phosphorylation states of P20.

To validate this hypothesis, we established stable transfectants of L6 cells that overexpress P20 (Fig. 7A). GLUT4myc was also coexpressed in these transfectants to increase insulin sensitivity (27). In the myotube cells overexpressing GLUT4myc alone, 50 nmol/l insulin increased 2-deoxyglucose uptake by 2.94- ± 0.31-fold over the basal level (Fig. 7B). This insulin-stimulated glucose uptake was decreased by 28% in the presence of 50 nmol/l amylin. However, in cells overexpressing both P20 and GLUT4myc, insulin-stimulated glucose uptake was significantly decreased by 41 ± 3% (n = 4; P < 0.05), whereas the inhibitory effect of amylin was significantly increased by 24 ± 2% (n = 4; P < 0.05). These results demonstrated that overexpression of P20 suppresses insulin-stimulated glucose uptake and enhances amylin’s ability to inhibit insulin’s action in L6 myotubes, thus suggesting a direct role of this protein in the regulation of glucose metabolism. The cellular mechanisms underlying such a regulatory role of P20 are currently under investigation in our laboratory.

In summary, our results demonstrated that insulin resistance in skeletal muscle is associated with the appearance of the two P20 phosphoisoforms, S2 and S3, and also with the inability of insulin to suppress the phosphorylation of these two isoforms. The role of P20 in the signal transduction pathways of insulin and amylin in skeletal muscle and the significance of the alterations in phosphorylation of this protein in insulin resistance remain to be clarified. Nevertheless, working backward toward receptors from phosphorylation of the three isoforms of P20 could serve as a method for elucidating postreceptor events of amylin and insulin and the functional interactions between these two hormones. Further studies on the intracellular defect that leads to the alteration of P20 phosphorylation in the insulin-resistant state could also help in understanding the pathogenesis of this disease.

These studies were supported by grants from the Endocore Research Trust and the Health Research Council of New Zealand.

We thank Dr. David James for kindly providing the pCXN2-GLUT4myc vector and Dr. Bernard Choong, Dr. Anthony Phillips, and Tom Mulvey for expert technical assistance.