The zinc transporter Zip14 (SLC39a14) affects Beta-cell Function: Proteomics, Gene expression, and Insulin secretion studies in INS-1E cells

Insulin secretion from pancreatic beta-cells is dependent on zinc ions as essential components of insulin crystals, zinc transporters are thus involved in the insulin secretory process. Zip14 (SLC39a14) is a zinc importing protein that has an important role in glucose homeostasis. Zip14 knockout mice display hyperinsulinemia and impaired insulin secretion in high glucose conditions. Endocrine roles for Zip14 have been established in adipocytes and hepatocytes, but not yet confirmed in beta-cells. In this study, we investigated the role of Zip14 in the INS-1E beta-cell line. Zip14 mRNA was upregulated during high glucose stimulation and Zip14 silencing led to increased intracellular insulin content. Large-scale proteomics showed that Zip14 silencing down-regulated ribosomal mitochondrial proteins, many metal-binding proteins, and others involved in oxidative phosphorylation and insulin secretion. Furthermore, proliferation marker Mki67 was down-regulated in Zip14 siRNA-treated cells. In conclusion, Zip14 gene expression is glucose sensitive and silencing of Zip14 directly affects insulin processing in INS-1E beta-cells. A link between Zip14 and ribosomal mitochondrial proteins suggests altered mitochondrial RNA translation, which could disturb mitochondrial function and thereby insulin secretion. This highlights a role for Zip14 in beta-cell functioning and suggests Zip14 as a future pharmacological target in the treatment of beta-cell dysfunction.


Supplementary 1. Flow cytometric gating strategy for experiments using PrimeFlow RNA Assays. Gating was performed by excluding debris (A) and cells aggregated in both forward-and side-scatter dimensions (B and C), and by eliminating dead cells (D). Example of insulin (Ins)
mRNA signal after gating (E). Fluorescence minus one (FMO) controls were included (F, Zip14; G, Znt8; H, insulin). Light gray histograms correspond to the indicated mRNA probe, while dark gray histograms are the corresponding FMO controls. One representative example following incubation of cells in 11 mM glucose is shown. Three hundred thousand cells were collected per sample. Large-scale proteomic analysis: The regulation of cellular proteins was evaluated following Zip14 silencing using a large-scale proteomic approach. The following procedure was used.
Tandem Mass Tag (TMT) Labeling and peptide purification: Seventy micrograms of protein was used, and Zip14 siRNA-and control siRNA-treated cells were compared (n = 4 replicates). An isobaric tag was added to each sample using a TMT 10plex Mass Tag Labeling Kit (Applied Biosystems, Foster City, CA, USA), according to the manufacturer's protocol. Briefly, samples were precipitated with ice-cold acetone, followed by protein digestion with trypsin overnight at 37°C. Samples were labeled with different TMT label reagents, included in the TMT Mass Tag Labeling Kit, and combined into one sample. They were then purified using strong cation exchange chromatography, followed by separation of peptides by isoelectric focusing on an Immobiline DryStrip (pH 3-10, GE Health Care Life Sciences, Uppsala, Sweden) using a Multiphor II unit (Pharmacia Biotech AB, Uppsala, Sweden). The DryStrip was cut into 10 pieces, and the peptides were extracted using AcN and trifluoracetic acid. Peptides were purified using C18-LC (Pierce C18 Spin columns), as recommended by the manufacturer, and evaporated before further analysis.
Nano-liquid chromatography and mass spectrometry analysis: The peptide mixtures were separated by nano-LC (Ultimate 3000, Dionex) coupled to a mass spectrometer (Orbitrap Fusion, Thermo Fisher Scientific, Bremen, Germany) through an EASY-Spray nano-electrospray ion source (Thermo Scientific). A µ-Precolumn (300 µm × 5 mm, C18 PepMap100, 5 µm, 100 Å, Thermo Scientific) and an analytical column (EASY-Spray Column, 500 mm × 75 µm, PepMap RSCL, C18, 2 mm, 100 Å, Thermo Scientific) were used to trap and separate the peptides, respectively. The peptides were eluted with a flow of 300 nl/min using a 125 min gradient by mixing Buffer A (0.1% FA) with Buffer B (80% AcN, 20% H2O, 0.1% FA). The gradient steps were performed with the following amounts of Buffer B: 6% (at 0 min), 12% (3 min), 28% (88 min), 40% (95 min), 90% (100 min), 90% (110 min), 6% (111 min), and 6% (125 min). The MS detection consisted of a full Orbitrap scan (m/z 380-1500) at a resolution of 120,000, with an automatic gain control (AGC) target of 2 × 10 5 and a maximum injection time of 50 ms. Up to 10 data-dependent MS 2 scans were performed in the linear ion trap in the mass range 400-1200 m/z, with CID energy at 35%, an AGC target of 1 × 10 4 , and a maximum injection time of 50 ms. The precursor ions were isolated using the quadrupole, with an isolation window of 0.7 m/z. The reporter ions were isolated with a window of 2 m/z and detected with data-dependent MS 3 using synchronous precursor selection in the Orbitrap in the mass range 120-500 m/z, with a HCD collision energy of 65%, and acquired at a resolution of 60,000, with an AGC target of 1 × 10 5 and a maximum injection time of 120 ms. Dynamic exclusion was set to 30 s or 70 s, in first and second replicate analysis, respectively.
Database searches and statistics: The raw MS data files from the two separate TMT studies were processed, and all generated peak lists from the same study were merged and analyzed using Mascot v. 2.5.1 (Matrix Science, London, UK) in Proteome Discoverer 2.1 (Thermo Scientific). In Proteome Discoverer 2.1, the following default workflows were applied: "Processing workflow for SPS MS3 reporter ion-based quantification" and "PWF_Fusion_Reporter_Based_Quan_SPS_MS3 _Mascot_Percolator". The MS data were searched against the SwissProt Rattus database downloaded in May 2016, which contained 25,741 proteins (including unreviewed sequences). Mascot was used for protein identification. Full scan tolerance was 8 ppm, MS/MS tolerance was 30 mmu, and up to two missed cleavages were accepted. Oxidation on methionine was set as a dynamic modification, while TMT-10-plex on lysine and N-terminal, and carbamidomethyl on cysteine, were chosen as static modifications. Only proteinunique peptides (Unique+Razor) and those consisting of at least six amino acids were included. Normalization to the summed intensities of the TMT signal was applied to compensate for possible variation in the starting material. The false discovery rate (FDR) criterion for peptides was q < 0.01. Only proteins with an identification score ≥ 30 (corresponding to a protein identification significance of 0.001) were considered, as well as proteins with at least three quantitative scans and one protein unique peptide, yielding 3431 proteins.
Mean protein ratios were calculated for each protein as the mean abundance in the four Zip14 siRNA-treated samples divided by the mean abundance in control siRNA-treated samples. A significant fold change was defined as 2× global standard error of the 3431 proteins (2 × 0.068 = 0.136). FDR was calculated from p-values (estimated by Student's t-test) using Benjamini and Hochberg's method [2]. A FDR < 0.05 was used. Fifty-two proteins met the fold-change and FDR criteria. For gene ontology analysis, to obtain protein groups, wider criteria, defined as a p value < 0.05 and a significant fold change, were used, yielding 121 proteins. For ontology analysis, the Functional Annotation tool from the Database for Annotation Visualization and Integrated Discovery (DAVID) was used (DAVID 6.8 Beta; https://david-d.ncifcrf.gov/). The standard settings of DAVID, with the addition of protein interactions, were applied with the stringency set as medium. The total list of the 3431 proteins found within the samples was selected as background. An enrichment score (ES) ≥ 1.3, equal to 0.05 on a non-log scale, was set as the level of significance [3]. Three protein lists were analyzed for functional enrichment; one included both up-and downregulated proteins and two consisted of only either down-or upregulated proteins.

Supplementary Material 4.
The full list of identified proteins in INS-1E cells by proteomics.