Structure-functional changes in eNAMPT at high concentrations mediate mouse and human beta cell dysfunction in type 2 diabetes

Aims/hypothesis Progressive decline in functional beta cell mass is central to the development of type 2 diabetes. Elevated serum levels of extracellular nicotinamide phosphoribosyltransferase (eNAMPT) are associated with beta cell failure in type 2 diabetes and eNAMPT immuno-neutralisation improves glucose tolerance in mouse models of diabetes. Despite this, the effects of eNAMPT on functional beta cell mass are poorly elucidated, with some studies having separately reported beta cell-protective effects of eNAMPT. eNAMPT exists in structurally and functionally distinct monomeric and dimeric forms. Dimerisation is essential for the NAD-biosynthetic capacity of NAMPT. Monomeric eNAMPT does not possess NAD-biosynthetic capacity and may exert distinct NAD-independent effects. This study aimed to fully characterise the structure-functional effects of eNAMPT on pancreatic beta cell functional mass and to relate these to beta cell failure in type 2 diabetes. Methods CD-1 mice and serum from obese humans who were without diabetes, with impaired fasting glucose (IFG) or with type 2 diabetes (from the Body Fat, Surgery and Hormone [BodyFatS&H] study) or with or at risk of developing type 2 diabetes (from the VaSera trial) were used in this study. We generated recombinant wild-type and monomeric eNAMPT to explore the effects of eNAMPT on functional beta cell mass in isolated mouse and human islets. Beta cell function was determined by static and dynamic insulin secretion and intracellular calcium microfluorimetry. NAD-biosynthetic capacity of eNAMPT was assessed by colorimetric and fluorescent assays and by native mass spectrometry. Islet cell number was determined by immunohistochemical staining for insulin, glucagon and somatostatin, with islet apoptosis determined by caspase 3/7 activity. Markers of inflammation and beta cell identity were determined by quantitative reverse transcription PCR. Total, monomeric and dimeric eNAMPT and nicotinamide mononucleotide (NMN) were evaluated by ELISA, western blot and fluorometric assay using serum from non-diabetic, glucose intolerant and type 2 diabetic individuals. Results eNAMPT exerts bimodal and concentration- and structure-functional-dependent effects on beta cell functional mass. At low physiological concentrations (~1 ng/ml), as seen in serum from humans without diabetes, eNAMPT enhances beta cell function through NAD-dependent mechanisms, consistent with eNAMPT being present as a dimer. However, as eNAMPT concentrations rise to ~5 ng/ml, as in type 2 diabetes, eNAMPT begins to adopt a monomeric form and mediates beta cell dysfunction, reduced beta cell identity and number, increased alpha cell number and increased apoptosis, through NAD-independent proinflammatory mechanisms. Conclusions/interpretation We have characterised a novel mechanism of beta cell dysfunction in type 2 diabetes. At low physiological levels, eNAMPT exists in dimer form and maintains beta cell function and identity through NAD-dependent mechanisms. However, as eNAMPT levels rise, as in type 2 diabetes, structure-functional changes occur resulting in marked elevation of monomeric eNAMPT, which induces a diabetic phenotype in pancreatic islets. Strategies to selectively target monomeric eNAMPT could represent promising therapeutic strategies for the treatment of type 2 diabetes. Electronic supplementary material The online version of this article (10.1007/s00125-019-05029-y) contains peer-reviewed but unedited supplementary material, which is available to authorised users.


Native Mass Spectrometry
The protein was expressed and isolated as previously described and, on the day of analysis, the buffer was exchanged into 100 mM ammonium acetate (Fisher Scientific, Loughborough, UK) pH 6.9 using micro Bio-Spin Chromatography columns (Micro Bio-Spin 6 Columns, Bio-Rad, Watford, UK) following the instructions specified by the manufacturer. The procedure was repeated twice and diluted to give a final concentration of WT NAMPT (5 mol/l) or S200D NAMPT (5 mol/l). Native MS data was acquired on the Synapt G2S HDMS (Waters, Manchester, UK). NanoESI capillaries were prepared in-house from thin-walled borosilicate capillaries (inner diameter 0.9 mm, outer diameter 1.2 mm, World Precision Instruments, Stevenage, UK) using a Flaming/Brown P-1000 micropipette puller (Sutter Instrument Company, Novato, CA, USA). A positive voltage was applied to the solution via a platinum wire (Goodfellow Cambridge Ldt, Huntington, UK) inserted into the capillary. Gentle source conditions were applied to preserve the native-like structure: capillary voltage 1.2-1.5 kV, sampling cone 50-200 V, source temperature 70 °C. Trap collision energy was 4 V, transfer collsion energy was set to 0 V. Nitrogen was the carrier gas. External calibration of the spectra was achieved using solutions of cesium iodide (2 mg/mL in 50:50 water:isopropanol). Data were acquired and processed with MassLynx software (Waters, Manchester, UK).

eNAMPT protein generation
Construction of DNA plasmids: DNA for full length mouse NAMPT was synthesised as a double stranded DNA gBlock (Integrated DNA Technologies) and cloned into pET151 according to

Purification of eNAMPT-WT and eNAMPT-WT (SSDD) mutation: Pellets were thawed and EDTA
Complete protease inhibitor tablets added (Roche, Welwyn Garden City, UK) then the E. coli were lysed using BugBuster (Merck Millipore, Burlington, MA, USA) according to the manufacturer's instructions. The NAMPT was found in the soluble fraction and was purified by passing over a 1mL Histrap ff crude column (GE Healthcare, Chicago, IL, USA) using a Biorad NGC system. The column was washed with 30 column volumes of 10mM sodium phosphate, 500 mmol/l NaCl, 20 mmol/l imidazole pH 7.4 and the bound protein eluted with 10 mmol/l sodium phosphate, 500 mmol/l NaCl, 500 mmol/l imidazole pH 7.4. Fractions containing protein were then pooled and concentrated (Amicon Ultra 15, Merck Millipore), then further purified using size exclusion chromatography.
Using a Gilson HPLC system, samples were run on a Superdex 200 increase 10/300 GL column in Phosphate buffered saline pH 7.4 (OXOID). Fractions corresponding to monomer and dimer were pooled individually. Final yield of protein was approximately 40mg/L for the WT, 8mg/L for monomer SSDD fraction and 5mg/L dimer SSDD fraction.

Pancreatic islet isolation
Mouse CD1 Islets were isolated as described previously [2]. Pancreata were inflated with 1mg/mL collagenase solution (Sigma-Aldrich, Poole, U.K.) followed by density gradient separation (Histopaque-1077; Sigma-Aldrich). Human islets were isolated from heart-beating non-diabetic donors, with appropriate ethical approval, at the King's College Hospital Human Islet Isolation Unit [2]. Isolated islets were incubated overnight (37°C, 5% CO2) prior to treatments. See human islet checklist table for additional details of human islet preparations.

Static and dynamic glucose-stimulated insulin secretion
For static insulin secretion, mouse CD1 or human islets were pre-incubated in a physiological salt solution [3] containing 2mM glucose. Groups of 3-5 size-matched islets were then further incubated at 37°C for 1 h in salt solution and 2 or 20 mmol/l glucose. Dynamic insulin secretion was measured using a temperature-controlled perifusion system. Mouse CD1 islets were transferred into chambers containing 1 μm pore-size nylon filters (40 islets/chamber) and perifused with a physiological salt solution [3] (37°C, 0.5mL/min) containing 2 or 20 mmol/l glucose or 20 mmol/l glucose and 20 mmol/l KCl. Samples were collected at 2-minute intervals throughout the experiment. Secreted insulin was measured using an in-house I 125 radioimmunoassay [3].

Quantitative RT-PCR
Total mouse CD1 or human islet RNA was extracted using Trizol reagent (Invitrogen, Paisley, UK).
Reverse transcription was performed using the High-Capacity cDNA reverse transcription kit (Applied Biosystems, Warrington, UK). Real-time qPCR was carried out with a LightCycler480, using Sybr Green PCR master mix (Qiagen, Hilden, Germany). Gene expression was measured by ΔΔCt methodology, normalised against GAPDH (Quantitech, UK). For primer details (all Eurogentec, Southampton, UK) see ESM Table 1.

NMN measurements
NMN was measured in MIN6 cells and human serum using a fluorometric assay based on previous described methodology [4,5]. protein was equalized to 10 g and an equal volume (30 l) was added to each well.

Mouse primers
Gene