Human neuroblastoma SK-N-MC cells (ATCC, Bethesda, MD, USA) were maintained in Minimum Essential Media (MEM, GIBCO-BRL, Gaithersburg, MD, USA) supplemented with 10% fetal bovine serum (FBS), 100 units/ml penicillin, 100 μg/ml streptomycin, and 2 mM L-glutamine at 37°C and 5% CO₂. The cells were treated with 2.5 mM of Aβ 1-42 for 24 h, to induce insulin resistance (Kornelius et al., 2017). Aβ 1-42 was acquired from AnaSpec Inc. (San Jose, CA, USA), dissolved in 100% 1,1,1,3,3,3-hexafluoro-2-propanol, and then dried using a vacuum desiccator. Next, Aβ was resuspended in dimethylsulfoxide (DMSO) and stored at -20°C (Kornelius et al., 2017). Icariin (I1286) was obtained from Sigma-Aldrich, Corp. (St. Louis, MO, USA) and dissolved in DMSO. The final concentration of DMSO in the cell supernatant was 1% (v/v). To measure insulin signaling and glucose uptake, the cells were serum-starved for 6 h and then stimulated with 100 nM of insulin for 10 min.

PTEN siRNA (#6538) and control siRNA (#6201) were obtained from Cell Signaling Technology (Beverly, MA, USA). Transfection of siRNAs was performed according to the manufacturer’s protocol. Western blot analysis was performed to measure the transfection efficiency.

RT-qPCR was performed as described previously (Li et al., 2018). The primer sequences used were as follows: PTEN forward, 5′-GTCACTGCTTGTTGTTTGC-3′ and reverse, 5′-TTCTTTGTTGATAGCCTCCAC-3′; U6 forward, 5′-GCTTCGGCAGCACATATACTAAAAT-3′ and reverse, 5′-CGCTTCACGAATTTGCGTGTCAT-3′. Relative gene expression was quantified by the 2^-ΔΔCt method and normalized to U6. All samples were analyzed in triplicates.

Adenovirus carrying the human PTEN gene (Ad/PTEN) was constructed as described previously (Stewart et al., 2002). Adenoviruses encoding β-galactosidase (Ad/β-gal) were kindly provided by Dr. Xuejun Wang (The University of South Dakota Sanford School of Medicine, SD, USA). The cells were incubated in serum-free medium containing adenoviruses for 6 h, and then grown in growth medium for another 36 h. Forty-two hours after virus infection, the cells were pending for experimental study. Ad/β-gal served as a negative control. The optimal multiplicity of infection (MOI) used in the experiments was 50 (Zhang et al., 2015).

The chymotrypsin-like activity of the 20S proteasome in the cells was measured as described previously (Zhang et al., 2015; Li et al., 2018). Briefly, the cells were lysed with cytosolic extraction buffer (50 mM HEPES, pH 7.5, 20 mM KCl, 5 mM MgCl₂, 2 mM ATP, 1 mM DTT, 0.025% Digitonin). The lysate was centrifuged at 10,000 × g for 10 min at 4°C, and the supernatants were collected. Protein concentration was determined through the bicinchoninic acid (BCA) assay. The synthetic fluorogenic peptide substrate Suc-LLVY-Amc (Boston Biochem, Cambridge, MA, USA) was used for assaying chymotrypsin-like peptidase activity. For assay specificity, 1 μM of proteasome inhibitor MG132 (S2619, Selleck, Shanghai, CN) was incubated with the extract. After 90 min of incubation at 37°C, the fluorescence intensity was read (excitation, 350 nm; emission, 440 nm) using fluorescence spectrometer (Perkin Elmer precisely LS 55, Billerica, MA, USA).

The Glucose Uptake Assay Kit (Colorimetric) (ab136955, Abcam Inc., Cambridge, MA, USA) was used to measure glucose uptake according to the manufacturer’s protocol.

Quantification of proteins or phosphorylated proteins was performed by western blotting, as described previously (Zhang et al., 2015; Li et al., 2018). Briefly, the cells were lysed in lysis buffer (50 mM HEPES, pH 7.6, 150 mM NaCl, 1% Triton X-100, 10 mM NaF, 20 mM sodium pyrophosphate, 20 mM β-glycerol phosphate, 1 mM sodium orthovanadate, 10 μg/ml leupeptin, 10 μg/ml aprotinin, and 1 mM phenylmethanesulfonyl fluoride). The lysates were incubated on ice for 20 min and then centrifuged at 14,000 × g for 10 min at 4°C. The supernatants were mixed with an equal volume of 2 x SDS-PAGE loading buffer and then heated at 95°C for 10 min. The proteins were separated by SDS-PAGE, transferred to a nitrocellulose membrane, and detected with specific antibodies.

The data are presented as the mean ± standard deviation (SD). The differences between the groups were examined for statistical significance using analysis of variance (ANOVA), followed by the Newman–Keuls post hoc test. A p-value <0.05 was considered statistically significant. The figures are the representative of at least four independent experiments with similar results.

To observe the potential effects of icariin on neuronal insulin resistance, human neuroblastoma SK-N-MC cells were serum-starved for 6 h and then treated with or without 2.5 mM of Aβ 1-42, in the presence or absence of 25, 50, or 100 μM of icariin for 24 h, followed by stimulation with or without 100 nM of insulin for 10 min. Aβ was used to induce neuronal insulin resistance in vitro (Zhao et al., 2008; Najem et al., 2016; Sajan et al., 2016; Wani et al., 2019) and insulin was administered to stimulate insulin signaling. As shown in Figure 1, icariin treatment reversed the inhibitory effects of Aβ on insulin-stimulated phosphorylation of Akt at Thr308 and of its downstream substrate AS160 (Figures 1A, B). To test the effects of icariin and Aβ on glucose uptake, SK-N-MC cells were serum-starved for 6 h and then treated with or without 2.5 mM of Aβ 1-42, in the presence or absence of 50 μM of icariin for 24 h, followed by stimulation with or without 100 nM of insulin for 10 min. We found that icariin treatment significantly increased 2-deoxy-D-glucose (2-DG) uptake (Figure 1C) under stimulation with insulin, when compared with Aβ treatment alone. These results suggested that icariin treatment protected SK-N-MC cells against Aβ-induced insulin resistance.

To determine the mechanisms underlying the protective effects of icariin on Aβ-induced insulin resistance, the expression levels of key proteins involved in the regulation of Akt phosphorylation, such as IRS-1, PTEN, PDK1, and TRB3, were investigated. SK-N-MC cells were serum-starved for 6 h and then treated with or without 2.5 mM of Aβ 1-42 for 24 h. We found that Aβ 1-42 treatment significantly enhanced only PTEN protein levels (Figures 2A, B) whereas PTEN mRNA levels remained unchanged (Figure 2C). To confirm the role of PTEN in Aβ-induced insulin resistance, PTEN was knocked down (KD) in SK-N-MC cells by siRNA techniques. Wild-type or PTEN KD cells were treated with or without 2.5 mM of Aβ 1-42 for 24 h, followed by stimulation with or without 100 nM of insulin for 10 min. We found that PTEN depletion markedly prevented Aβ-induced reduction of insulin-stimulated Akt T308 phosphorylation (Figures 2D, E) and glucose uptake (Figure 2F). These results suggested that the increased PTEN levels were responsible for Aβ-induced neuronal insulin resistance in vitro.

To check the effects of icariin on PTEN, SK-N-MC cells were serum-starved for 6 h and then treated with or without 2.5 mM of Aβ 1-42, in the presence or absence of 50 μM of icariin for 24 h. We found that icariin treatment significantly reduced Aβ-enhanced PTEN protein levels (Figures 3A, B). To investigate whether PTEN is a target of icariin’s activity, a PTEN rescue experiment was performed, in which PTEN was overexpressed by infecting the cells with Ad/PTEN. Ad/β-gal acted as a negative control. The control or PTEN-overexpressing SK-N-MC cells were serum-starved for 6 h and then treated with or without 2.5 mM of Aβ 1-42, in the presence or absence of 50 μM of icariin for 24 h, followed by stimulation with 100 nM of insulin for 10 min. We found that the PTEN-overexpressing SK-N-MC cells displayed a reduction in insulin-stimulated Akt T308 phosphorylation (Figures 3C, D) and glucose uptake (Figure 3E) in response to the combined treatment with Aβ and icariin, when compared with the control cells.

In addition, the results of the RT-qPCR revealed that icariin did not affect the mRNA levels of PTEN (Figure 4A, Figure S1). Interestingly, Aβ treatment significantly inhibited proteasome activity, which was prevented by icariin administration (Figure 4B). In line with this finding, we also observed that icariin administration reduced the Aβ-enhanced ubiquitination of total proteins (Figure 4C). These findings suggested that icariin reduced PTEN protein levels via a proteasome-dependent mechanism. To confirm this hypothesis, SK-N-MC cells were serum-starved for 6 h, pretreated with 2 μM of proteasome inhibitor MG132 for 30 min, and then treated with or without 2.5 mM of Aβ 1-42, in the presence or absence of 50 μM of icariin for 24 h. We found that MG132 administration significantly increased PTEN protein levels, when compared with the group that received a combined treatment of Aβ and icariin (Figures 4D, E).

Taken together, these results suggested that icariin activated the proteasome, leading to a reduction of PTEN protein levels, by which icariin exhibited its protective effects against Aβ-induced neuronal insulin resistance.

In the present study, we described the molecular mechanisms by which icariin mitigated Aβ-induced neuronal insulin resistance. However, the majority of the results were based on Aβ-treated SK-N-MC cells, which represent an in vitro model of uncertain relevance to AD. In addition, the insulin concentration used in the performed experiments was higher than the physiological one. The concentrations of icariin were also relatively high. Therefore, further studies are necessary to obtain more potent derivatives of icariin, as well as to provide further in vivo evidence in experimental settings closer to clinical reality.