Mouse models
All animal procedures were approved by the Institutional Animal Care and Use Committee at UConn Health adhering to federal and state laws. Mice with the exon 5 of Ubxn9 flanked by loxP sites (Ubxn9flox/flox) were generated via homologous recombination. The homozygous Ubxn9flox/flox were then crossed with homozygous tamoxifen-inducible Cre recombinase-estrogen receptor T2 mice (The Jackson Laboratory, Stock # 008463; Rosa26Cre-ERT2) to generate male and female Cre+/-Ubxn9fl littermates, which were mated to produce Cre+/-Ubxn9bfl/fl, Ubxn9bfl/fl, and Cre+/-. To induce global Ubxn9 knockout in ≥ 5-week-old mice (Ubxn9-/-), 1mg of tamoxifen was administered (dissolved in corn oil) to each mouse every other day for a total of 5 injections. ERT2-Cre+/-Ubxn9fl/fl treated with corn oil served as the wild-type control (Ubxn9+/+). Tamoxifen was allowed to be cleared 10 days after the last injection before experimentation. Both male and female mice were used between 7–18 weeks of age. Genotyping was performed with genomic DNA and Choice Taq Blue Mastermix (Denville Scientific, Cat# CB4065-8) using the following PCR protocol: 95°C for 1 s, 34 cycles of 94°C for 1 min, 60°C for 30 s, 72°C for 30 s, and then 72°C for 7 min, 4°C to stop. The genotyping primers for the loxP sites in exon 5 were: Ubxn9 F 5’GCTTCTCTCAAAGCTGGAGAGTCAC; R 5’ CAAGGCACTGGGCCAGGGAG. The PCR reaction resulted in a product of 226bp (WT) and/or 276 bp (loxP). The Cre primers were common: WT F 5’-AAGGGAGCTGCAGTGGAGTA; WT R 5’- CCGAAAATCTGTGGGAAGTC; mutant R 5’-CGGTTATTCAACTTGCACCA. The PCR reaction resulted in a product of 297 bp (WT) and/or 450bp (Cre).
The GLUT4 LoxP mice (Slc2a4flox/flox) were generated as previously described 71 and provided by Carol A. Witczak (Indiana University). Slc2a4fl/fl mice were then bread with muscle creatine kinase Cre recombinase transgenic mice (The Jackson Laboratory, strain # 006475; MCK-Cre+). Heterozygote mice (Slc2a4fl/+-MCKCre) were used for breeding to generate the following genotypes: wild-type (Slc2a4+/+), GLUT4 LoxP HET (Slc2a4fl/+), LoxP homozygous (Slc2a4fl/fl), MCK-Cre+ (control), muscle-specific GLUT4 heterozygous (Slc2a4fl/+ MCKCre) and muscle-specific GLUT4 knockout (KO; Slc2a4fl/fl MCKCre). For primary myocyte studies, Slc2a4fl/fl MCKCre (KO) and MCK-Cre+ (WT) mice were used at 12-weeks of age. For genotyping, the PCR protocol for GLUT4 LoxP was 95oC for 3min, 35 cycles of 95oC for 20sec, 64oC for 30sec, 72oC for 50sec, 72oC for 2min, 10oC to stop. The genotyping primers for LoxP were F: 5'GGCTGTGCCATCTTGATGACC 3’; R: 5'ACCCATGCCGACAATGAAGTTAC 3’. This PCR resulted in a WT band (~ 752 bp) and a GLUT4 LoxP + band (~ 900bp). The Cre primers for MCK-Cre were F: 5’TAAGTCTGAACCCGGTCTGC 3’; R: 5’GTGAAACAGCATTGCTGTCACTT 3’ and resulted in Cre + band (450bp). An internal control primer for DNA quality was also included F: 5'CAAATGTTGCTT GTCTGGTG 3’; R: 5'GTCAGTCGAGTGCACAGTTT 3’ and resulted in a band ~ 200bp. The Cre PCR protocol was 94oC for 2min, stepdown protocol of 0.5oC/cycle at 94oC for 1min, 65oC 30sec, 68oC 50sec, then 28 cycles of 94oC 1min, 60oC for 30sec, 72oC for 50sec, then 72oC for 7min, 10oC hold (The Jackson Lab strain # 006475; MCK-Cre+, Cre primer protocol).
Reagents and antibodies
The anti-GAPDH (Cat # 60004-1-Ig, 1:2000) and anti-GFP (Cat # 50430-2-AP, 1:100 for IF) were purchased from Proteintech Group, Inc (Rosemont, IL). The mouse anti-Myc (Cat # TA150121, 1:1000) and mouse Slc2a4 siRNA duplex (Cat # SR416343, 10nM final concentration) were obtained from Origene Technologies, Inc (Rockville, MD). The rabbit anti-GLUT4 (Cat # PA5-23052, 1:1000), rabbit anti-MDA5 (Cat # 700360, 5µg/mL for IF) Hoescht 33342 solution (Cat # 62249, 1:10,000 for IF), Pierce Protein A/G Magnetic beads (Cat # 88802), Pierce Anti-Myc Magnetic Beads (Cat # 88842) were bought from Thermo Fisher Scientific (Waltham, MA). To stain cell plasma membranes for confocal microscopy images, CellBrite™ Fix 555 (Cat # 30088-T, 1:1000 for IF) and CF® 405M (Phalloidin conjugate; Cat # 00034-T, 1:200 for IF) was obtained from Biotium (Fremont, CA) and mouse anti-ZO-1 from Thermo Fisher Scientific (Cat #33–0100). GLUT4 antibodies (Clone LM048, 10µg/mL for IF) used for detecting endogenous mouse GLUT4 by immunofluorescence was purchased from Integral Molecular (Philadelphia, PA) 72. The anti-cardiac troponin T antibody (ab8295) used for cardiomyocyte verification was purchased from Abcam (Cambridge, United Kingdom). The mouse-anti FLAG antibody (Clone M2, Cat # F1804-200UG, 1:1000 for WB; 1:200 for IF), anti-FLAG M2 magnetic beads (Clone M2, Cat # M8823) and VSV-G antibody (Cat# V4888, 1:1000, Sigma-Aldrich) was purchased from Sigma-Aldrich (St. Louis, MO). The goat anti-RIG-I antibody (Cat # sc-48931, 1:100 for IF) was a product of Santa Cruz Biotechnology (Dallas, TX) and rabbit anti-RIG-I antibody (Cat # 700366, 5µg/mL for IF) was from Invitrogen (Waltham, MA). The anti-α tubulin (Cat # 3873S, 1:2000), Actin (Cat # 4970L, 1:2000), RIG-I (Cat # 3743S, 1:1000), MDA5 (Cat # 5321S, 1:1000), phosphorylated IRF3 (Cat # 4947P, 1:1000) and total IRF3 (Cat # 11904P, 1:1000), rabbit-anti mouse MAVS IP antibody (Cat # 83000S, 1:50), UBXN9 (Cat # 2049S, 1:1000), phosphorylated TBK1 (Cat # 5483S, 1:1000), ISG15 (Cat # 2743, 1:1000), LDHA/C (Cat # 3558S, 1:1000), Caveolin (Cat # 3267T, 1:1000), phosphorylated AKT (Cat # 4060T, 1:1000), total AKT (Cat # 9272, 1:1000), phosphorylated c-Cbl (Cat # 8869T, 1:1000), total c-Cbl (Cat # 2747S, 1:1000), phosphorylated AS160 (Cat # 8881, 1:1000), total AS160 (Cat # 2670, 1:1000), total AMPK (Cat # 2793, 1:1000), and phosphorylated AMPK (Cat # 2535, 1:1000) were all purchased from Cell Signaling Technologies (Danvers, MA). Antibody used to detect UBXN9 cleavage (directed against C-terminus) was described previously 17, 19. Immunofluorescence secondary antibodies including Goat anti-rabbit Alexa Fluor™ 594 (Cat # A11037, 1:200), Goat anti-human Alexa Fluor™ 488 (Cat # A11013, 1:200), Donkey anti-goat Alexa Fluor™ 594 (Cat # A11058, 1:200), Donkey anti-goat Alexa Fluor™ 488 (Cat # A11055, 1:200), Donkey anti-rabbit Alexa Fluor™ 488 (Cat # A21206, 1:200) and DAPI solution (Cat # D1306, 1:1000) were purchased from Thermo Fisher Scientific (Waltham, MA). TransIT-X2 dynamic delivery system (Cat # MIR6005) was obtained from Mirus Bio, LLC (Madison, WI). IFN stimulatory DNA derived from Listeria monocytogenes genome (ISD, Cat# tlrl-isdn), 5’ triphosphate hairpin RNA (Cat# tlrl-hprna), and high molecule weight polyinosine-poly cytidylic acid [HMW poly (I:C),1.5–8 kb, Cat# tlrl-pic] and mouse IFN-β ELISA kits were from Invivogen (San Diego, CA). The recombinant mouse IFN-β was from R&D systems (Cat # 824-MB-010/CF). Glucose uptake (2-NBDG, Item # 600470) and lactate concentrations (Cat # J5021) was quantified using kits from Cayman Chemical (Ann Arbor, MI) and Promega (Madison, WI), respectively. All primers used in this study were obtained from Integrated DNA Technologies, Inc. (Coralville, IA).
Isolation and culture of primary skeletal myoblasts and bone-marrow derived macrophages (BMDMs)
Primary skeletal myoblasts were isolated from Ubxn9+/+ and Ubxn9−/− mouse quadricep muscle tissue, purified and differentiated as previously described with minor adjustments 73. Quadriceps (vastus medialis and vastus lateralis) were excised from two hindlegs of mice and placed in 1X phosphate-buffered saline (PBS) for washing. Muscle was then finely minced into small pieces, transferred to a 15ml conical tube, and spun down at 21,130 x g for 30s at RT to collect muscle pieces. Muscle was digested in Collagenase II (Cat #: 17101015, Gibco™) digestion media (400 U/ml in DMEM -FBS, sterile filtered) on a shaker at 100 rpm for 1.5 hours with vortexing halfway through digestion time. After shaking, vortex for 15s and muscle mix should appear cloudy. Tubes were then spun down at 1,400 x g for 5 min at RT, at which time supernatant was removed and muscle was resuspended in complete DMEM to neutralize collagenase II. To release myoblasts from muscle, tissue was pipetted up and down 30X using a sterile 10ml pipette and then strained over a 70µm strainer on a 50 ml tube. Cells were collected and strained through a 40µm filter to remove macrophages. Tubes were spun down at 1,400 x g for 5 min at RT and myoblast pellet resuspended in myoblast growth medium (MGM: DMEM/F12 + 1% p/s, 20% FBS, 10ng/ml basic fibroblast growth factor). Cells were seeded on plates precoated with 10% Matrigel (Cat #: 354234, Corning™).
For purification of a pure myoblast culture, digested muscle/cells were allowed to attach to Matrigel-coated plates for 72h without media removal. Three-days later, small, droplet-shaped myoblasts and larger fibroblasts should be visualized. Here, cells are ready for first pre-plating step. MGM medium was removed, and cells washed 2X with MGM to remove lingering cell debris and trypsinized to detach cells. Cells were centrifuged at 800 x g for 5 min and plated in 6-well plates that were not coated with Matrigel for 1h. This step is critical to remove contaminating fibroblasts as these will attach to the uncoated surface and myoblasts will remain in suspension. After 1h, fibroblasts were visualized on plate surfaces and supernatant containing myoblasts was removed and added to a new 10% Matrigel-coated plate. These purification steps were performed every 48h until > 98% myoblasts were obtained (typically ~ 3 pre-plate steps are necessary). After a pure culture is achieved, cells are ready for experimentation, or alternatively, they were differentiated into myotubes using differentiation medium (DM: DMEM + 2% horse serum) once cells reached ~ 80–90% confluency. DM medium was changed every 48h until myoblasts align and fusion begins (~ 120h). At this point, cells were ready for experimentation. Skeletal muscle myoblasts were isolated from Slc2a4+/+ and Slc2a4−/− using the same protocol as above.
Bone marrow-derived macrophages were isolated and cultured from Ubxn9+/+ and Ubxn9−/− mice based on our previous studies 74. In brief, bone marrow cells were differentiated into macrophages (BMDM) in L929-conditioned medium (RPMI 1640, 20% FBS, 30% L929 culture medium, 1x antibiotics/antimycotics) in a 10cm Petri dish at 37°C, 5% CO2 for 5–7 days. The culture medium was replaced fresh every 2 days. Attached BMDMs were dislodged by pipetting or gentle scrapping and counted for plating in plating medium (RPMI 1640, 10% FBS, 5% L929-conditioned medium, 1x antibiotics/antimycotics).
Human iPSC culture and cardiomyocyte differentiation
Human iPSCs were generated as previously described 75. Briefly, the fibroblasts were derived from discarded female neonatal skin tissue under Yale Institutional Review Board approval. CytoTuneTM-iPSC 2.0 Sendai Reprogramming Kit (ThermoFisher Scientific, A16517) was used to reprogram fibroblast cells into iPSCs (Y6-iPSC). Twenty-four hours after viral transduction, infected cells were harvested and plated onto mitotically arrested MEF feeder layer with human iPSC medium (20% KSOR in DMEM/F12 medium supplemented with 10ng/ml bFGF), 1% non-essential amino acid (v/v), 2 mM L-Glutamine, 0.44 µM beta-Mercaptoethaol, and 1% Pen/Strep (v/v) (all from ThermoFisher Scientific, USA). The medium was changed every other day for three weeks. Colonies of iPSCs were then picked and expanded on MEF feeder layers for cardiac differentiation.
The iPSCs (Y6-iPSC) were grown to 70–80% confluency on MEFs and then dissociated using 1mg/mL Dispase II (Gibco,17105041) for 7–9 min. At which time, the iPSCs were further mechanically dissociated using a 5 mL pipette, collected, and centrifuged at 200 rpm for 4 min. The top supernatant was discarded to deplete the MEF cells, and the iPSC pellet resuspended in mTeSR™ 1 (Stemcell Technologies, 05850) containing 5µM ROCK inhibitor (Y-27632 dihydrochloride, Tocris Bioscience; 1244) and re-plated 1:1.5 on Matrigel-coated plates (Corning, diluted 1:60 in DMEM). Once reaching 80–95% confluency, cardiac differentiation was initiated using 20µM CHIR99021 (Selleckchem, S2924) in RPMI-1640 supplemented with 1% B27 minus insulin (Gibco, A18956-01) and one volume mTeSR™ 1 (considered day 1 of culture), as described in 28. On day three, medium was supplemented with 1% B27 minus insulin with the addition of 5µM IWP-4 (Stemgent Reprocell, 04–0036). After two days, the medium was switched to 1% B27 minus insulin without additional factors. Beating cardiomyocytes were typically observed on day 9–11. On day 11, medium was changed to 1% B27 containing insulin (Gibco, 17504044). Noncardiomyocytes were next eliminated starting from day 13 by treating cells with 4mM lactate (Sigma-Aldrich; L7022) in DMEM without glucose (Gibco; 11966025) for 4 days, with a medium change every other day. Enriched cardiomyocytes were then dissociated into clusters using a mixture of 10µg/mL Collagenase A and B (Roche, 10103586001 and 11088807001) for 30 min at 37oC. Cells were further dissociated into single cells using Accutase™ (Sigma-Aldrich; A6964) for 10 min at 37oC. Single cardiomyocytes were then plated onto fibronectin-coated plates. The RPMI1640 medium containing 1% B27 with insulin was replaced every other day for 2 weeks. On day 35, cardiomyocytes were processed for immunofluorescence (cardiac troponin T, cTnT) and siRNA/EMCV experiments.
Mouse infection and disease monitoring
Ubxn9 +/+ and Ubxn9−/− mice were intraperitoneally injected with 100 PFU of EMCV and morbidity and mortality were monitored twice a day for survival studies. Prior to infection, mice were fasted overnight to collect baseline IFN-β, glucose, lactate, and viral RNA levels (considered cutoff for the level of detection by qPCR). Mice were then allowed to feed ad libitum following infection. IFN-β in the plasma was analyzed at 24 h post-infection (hpi) from mice injected with 1000 PFU EMCV. Viral titers in plasma and heart tissue at 3 days post-infection (dpi) and 4dpi were assessed by qPCR and plaque assay, respectively. Lactate concentrations were quantified from the same samples used for qPCR of viral RNA and quantified using a Lactate-Glo Assay (Promega, Madison, WI, Cat # J5021).
Cell culture and viruses
EMCV (Cat # VR129-B) and VSV (Indiana Strain, Cat # VR-1238) were purchased from American Type Culture Collection (ATCC) (Manassas, VA) and the multiplicity of infection was specified in each figure legend. ONNV UgMP30 strain (NR-51661) was provided by BEI Resources. Green fluorescence protein (GFP)-VSV was made by inserting a VSV-G/GFP fusion sequence between the VSV G and L genes and propagated in our lab for use in several studies 25, 74, 76. These viruses were propagated in Vero cells and titrated by a plaque forming assay.
HEK293T cells (Cat# CRL-3216), Vero cells (monkey kidney epithelial cells, Cat# CCL-81), C2C12 (mouse muscle cells, Cat # CRL-1772) and A549 (human lung cell line, Cat #CCL-185) were obtained from ATCC (Manassas, VA). The 3T3-L1 Myc-GLUT4-GFP fibroblast/adipocyte cell line was generated in our studies previously 34, 77. All cell lines were grown in Dulbecco’s modified Eagle medium (DMEM, Life Technologies, Grand Island, NY) supplemented with 10% FBS fetal bovine serum (FBS) and antibiotics/antimycotics. These cell lines are not listed in the database of commonly misidentified cell lines maintained by ICLAC and have not been authenticated in our hands. They are routinely treated with MycoZAP (Lonza) and tested for mycoplasma contamination in our hands. Cells were maintained in between experiments using stock DMEM supplied by Life Technologies (Grand Island, NY). During experiments with virus, ligands, etc., cells were cultured in base DMEM without glucose, sodium pyruvate or HEPES (Thermo Fisher Scientific, Cat # 11966025) supplemented with 10% FBS. Glucose was then complemented back at the same concentration (25mM) normally present in DMEM. This was to account for sodium pyruvate that feeds lactate and would skew glucose uptake/glycolysis measurements. For bioassay, 2fTGH-ISRE reporter cells were used to determine the concentration of IFN-β in the cell supernatant of A549 cells 76.
For differentiation of myotubes from C2C12 myoblasts, cells were seeded at ~ 70% confluency the day before beginning protocol. When cells reached ~ 95–100% confluency, culture medium was replaced with DMEM + 2% horse serum (HS) and medium replaced every 48h for 5–6 days, or until most of cells were fused together. Experiments were then carried out on myotubes. Murine 3T3-L1 fibroblasts were allowed to reach confluency at least two days prior to induction of differentiation into adipocytes. Differentiation was induced on the first day with DMEM containing 0.25 µM dexamethasone, 160 nM insulin, and 500 µM methylisobutylxanthine, followed by feeding cells with fresh medium containing 160nM insulin for two more days. On day four, cells were maintained in DMEM + 10% FBS until 8 days. Experiments were then performed on differentiated adipocytes ~ 8–10 days after process was initiated.
Ligand treatments, siRNA, and in vitro virus infection
PRR ligands were transfected into cells with TransIT-X2 (Madison, WI, Product # MIR6005) using standard procedures described in product manual. In brief, DNA or RNA agonists were mixed with TransIT-X2 in 1X Opti-MEM (Gibco|Thermo Fisher Scientific, Waltham, MA, Ref # 31985-070) for 20 min before transfecting mixture dropwise onto cells. The concentrations of high-molecular-weight polyinosine-polycytidylic acid [HMW Poly(I:C)] (InvivoGen, San Diego, CA, Cat # tlrl-pic), 5’triphosphate (3p-hpRNA) (InvivoGen, San Diego, CA, Cat # tlrl-hprna-100) and interferon stimulatory DNA (ISD) (InvivoGen, San Diego, CA, Cat # tlrl-isdn) used for in vitro stimulation were specified for each experiment in the figure legend.
Knockdown of GLUT4 (Slc2a4) in C2C12 cells was done as described in user manual (Origene, Rockville, MD, Cat # SR416343). In brief, three unique targeting siRNA’s for Slc2a4 (final concentration: 50nM) were mixed with TransIT-X2 in 1X Opti-MEM for 20 min before transfecting mixture dropwise onto cells. After 48h of incubation to achieve knockdown, cells were then treated with 3p-hpRNA as described above. A universal, non-specific scramble siRNA was used as a non-targeting control. The same protocol was adopted for knockdown of Ubxn9 (Origene, Rockville, MD, Cat # SR417176). For iPSC-CMs, knockdown of human GLUT4 using two unique siRNA’s (final concentration: 50nM) were prepared based on the user manual (Origene, Rockville, MD, Cat # SR304402) and transfected as above.
For viral infection, viruses were diluted in DMEM without FBS and allowed to attach and infect cells for 2h; the cells were then washed with 1X phosphate-buffered saline (PBS) once and incubated with fresh medium. The MOI and infection time were denoted in each experimental figure legend.
Plaque-forming assay
Quantification of infectious EMCV (Cat# VR-129B) viral particles in heart tissue/cell culture supernatant was performed on Vero cell monolayer with minor modifications 74. Briefly, heart tissue was weighed and 15mg of heart was digested in PBS using a Bio-Gen PRO200 Homogenizer (Pro Scientific, Oxford, CT, Cat # 01-01200). A total of 30–100 µg (total proteins) of tissue lysate or cell supernatant was serially diluted in DMEM (-) FBS and applied to confluent Vero cells (12-well plate) at 37°C for 2h. The inoculum was then removed and replaced with 2 ml of DMEM complete medium with 1% SeaPlaque agarose (Lonza, Cat# 50100). Plates were inverted at 37°C, 5% CO2 and plaques visualized using Neutral red (Sigma-Aldrich) after 24h of incubation. Viral titers were expressed as plaque forming units (PFU) / mL or gram of tissue.
Plasmid construction and molecular cloning
The pB retrovirus vector containing GLUT4myc7-GFP (herein referred to as pGLUT4-GFP) was used in our studies previously 77. This plasmid was overexpressed in C2C12 using similar transfection methods described previously. The human full length (FL) and R169A point mutant GLUT4 were provided by Chuangye Yan at Tsinghua University 41. The FL (NCBI accession: NM_001042) and R169A variant were cloned into a new, custom pcDNA3.1-FLAG vector 74 for expression as an N-terminal FLAG-fusion protein using standard PCR amplification and cloning techniques. The pcDNA3.1-FLAG vector was also used to generate FLAG-UBXN9 76. To generate GLUT4 truncations on the same plasmid backbone, the FL GLUT4 plasmid served as a template to clone GLUT4 ΔN24, ΔL6, ΔC42 mutants (PCR primers from Integrated DNA Technologies, Coralville, IA) and these were subsequently cloned using the same protocol. The pcDNA-FLAG-GLUT4 plasmids were transformed into E. coli DH5α (Thermo Fisher Scientific, Cat # 18265017) bacteria by electroporation at 42oC for 45sec. E. coli were plated on LB Agar containing 100µg/mL ampicillin and inverted overnight at 37oC. Antibiotic-resistant colonies were picked and grown in LB media at 37oC overnight by shaking (100rpm). Plasmid DNA was extracted using PureLink™ Quick Plasmid Miniprep Kit (Thermo Fisher Scientific, Cat # K210011). Sanger sequencing was used to verify GLUT4 insertion into plasmids. RIG-I mutant plasmids were similarly cloned as above where the human FL RIG-I inserted into the pcDNA3.1-FLAG vector 74 served as the template to clone RIG-I CARD, Helicase, ΔCARD and CTD. MDA5 FL was also cloned into the pcDNA3.1-FLAG vector and served as the template to clone MDA5 CARD and ΔCARD plasmids.
Generation of gene knockout cell lines with CRISPR-Cas9 technology
Pre-designed, gene specific guide (g) RNAs (Integrated DNA Technologies, Coralville, IA) were subcloned into a lentiCRISPR-v2 vector 78 and correct insertion was confirmed by sequencing. To generate lentiviral particles, each gRNA vector was transfected into HEK293T cells with the packaging plasmids pCMV-VSV-G and psPAX2 (Didier Trono lab via Addgene, Watertown, MA, Cat # 12259). After 24h, half of the cell culture medium was replaced with DMEM, and viral particles were collected at 48-72h after transfection. The viral supernatant was cleared by brief centrifugation at 2000 rpm for 5min. C2C12 target cells were then seeded at ~ 50% and transduced the next day with each individual lentivirus. The WT control was lentiCRISPRv2 vector only. C2C12 cells were then selected with 0.8 µg/mL of puromycin for 10–20 days, changing puromycin media every other day. Successful knockout clones were confirmed by western blotting. Guides targeting mouse UBXN9 (F- CACCGATGAAGTGCTACGACCCCGT; R-5’ AAACACGGGGTCGTAGCACTTCATC) GLUT4 (F-5’ CACCGAGGCACCCTCACTACGCTCT; R-5’ AAACAGAGCGTAGTGAGGGTGCCTC), LDHA (F-5’ CACCGTGTTCACGTTTCGCTGGACC; R-5’ AAACGGTCCAGCGAAACGTGAACAC), and AKT2 (F-5’ CACCGTCACAAAGCATAGGCGGTCA; R- 5’ AAACTGACCGCCTATGCTTTGTGAC) were used in this study.
Purification of total cellular RNA, reverse transcription and quantitative (q) PCR
Approximately 15mg of mouse tissue, ~ 20 µL of whole blood and up to 1x 106 culture cells were collected in 300 µL of lysis buffer (RNApure Tissue & Cell Kit, CoWin Biosciences, Cambridge MA, United States). Heart tissue was homogenized as described, mixed with lysis buffer, and extracted according to the product manual. Isolated RNA was quantified using a spectrometer feature on BioTek Cytation 1 imaging reader (Agilent, Santa Clara, CA) and RNA concentration was normalized according to the lowest concentration across all samples with RNase-free water. RNA samples were normalized and converted into cDNA using the PrimeScript™ Reverse Transcription (RT) reagent Kit (TaKaRa Bio, Inc, Cat# RR037A). Quantitative PCR (qPCR) was performed with gene-specific primers and iTaq Universal SYBR Green Supermix (BioRad, Cat# 1725124). Results were calculated using the 2–ΔΔCt method from the CT values for each sample and gene of interest. A housekeeping gene (Actb, Polr2b) was used as an internal control. The qPCR primers have been used and validated in our previous studies 74, 79.
Mouse glucose tolerance test (GTT)
Ubxn9 +/+ and Ubxn9-/- mice were fasted overnight (~ 12h) followed by being administered with 1 mg glucose/kg body weight (Sigma, St. Louis, MO, Lot # 82H0725) via intraperitoneal (i.p.) injection. Tail vein blood at baseline and indicated time points after injection was collected and measured for glucose levels by using a handheld glucometer (Germaine Laboratories, San Antonio, TX). Glycemia was reported as mg/dl.
2-deoxyglucose (2-DG) and lactate assays
Cells were plated in 96-well clear bottom black plates one day before assay to allow equilibration in metabolism. To quantify 2-NBDG uptake following insulin treatment, protocol was followed according to manufacturer’s instructions, with minor modifications (Cayman Chemical, Ann Arbor, MI, Item # 600470). For acute glucose uptake, cells were serum starved in no glucose (-FBS) DMEM medium for 3h before 50 µg/ml 2-NBDG was added to no glucose DMEM containing FBS and 200nM human insulin (Millipore Sigma, Darmstadt, Germany, Cat # 91077C) for 15–20 min. Supernatant was removed and cells were washed 2X with assay buffer to remove non-specific binding. After second wash, cells were placed in assay buffer and intracellular 2-NBDG taken up by cells was detected with fluorescent filters (485nm/535nm) on BioTek Cytation 1 imaging reader (Agilent, Santa Clara, CA). To account for background signal, no insulin treated cells were used as “steady state” glucose consumption and 2-NBDG in DMEM was added to empty wells without cells. Both controls were subtracted from insulin treated groups to calculate relative glucose uptake. For experiments of longer duration (3p-hpRNA, virus, IFN-β), 2-NBDG was added to DMEM without glucose with or without specified treatment and incubated with cells until timepoint indicated in figure legends. To account for cell growth/background, “Mock” cells were incubated with 2-NBDG in DMEM alone and assayed in tandem with treated cells. Cells were washed and assayed as described above for insulin treatment. For testing inhibition of glucose uptake, cells were pretreated for 1h with 50µM of Fasentin (R&D Systems, Minneapolis, MN, Cat # 6100/10) diluted in DMEM without glucose (or DMSO control). Media was then replaced containing fresh inhibitor with 100 µg/ml 2-NBDG and cells were incubated for 1h before assessing glucose uptake as described previously.
Lactate was detected using a Lactate-Glo Assay (Promega, Madison, WI, Cat # J5021)
for extracellular supernatant (cell culture medium), intracellular (cell lysates) and mouse plasma. Protocol was followed exactly as suggested in product manual as each type of sample required unique processing. For intracellular lactate, cells were plated in a white 96-well clear bottom plates and treated as described in figure legends. After specified timepoints, supernatant was removed, and lysate used for determining intracellular lactate concentration. For mock groups, cells were incubated in media side-by-side with experimental wells and processed simultaneously. Mouse plasma was diluted 100-fold and processed as for cell culture supernatant samples. Plates were measured using a BioTek Cytation 1 imaging reader (Agilent, Santa Clara, CA) with luminescent filters. DMEM + 10% FBS was used as background control for cell culture supernatant and plasma; 1X PBS used for intracellular lactate background.
Enzyme-linked immunosorbent assay (ELISA) and immunoblotting
IFN-β in cell supernatants, homogenized tissue and plasma were assessed using LumiKine™ Xpress mIFN-β 2.0 ELISA detection kit according to manufacturer’s instructions (InvivoGen, San Diego, CA, Cat #luex-mifnbv2). Human IFN-β in cell supernatants of iPSCs were assessed using DuoSet® ELISA kit according to the manufacturer’s instructions (R&D Systems, Minneapolis, MN, Cat #DY814-05). The IFN-β data are presented as pg/ml. For western blotting analysis, cells were lysed in RIPA (Alfa Aesar, Tewksbury, MA, Cat # J63306) or for overexpression/detection of GLUT4, 2% n-Dodecyl-β-D-Maltopyranoside (Anatrace, Maumee, OH, Cat # D3101GM). For detection of GLUT4, samples were not boiled and lysed on a rotator for 2h at 4oC before centrifuging at 21,000 x g for 10min at 4oC. Samples were run on standard dodecyl sulfate-polyacrylamide gels and transferred onto nitrocellulose membranes using Mini Trans-Blot® Cell transfer systems (Biorad, Hercules, CA, Cat #1703930). Blots were then blocked with 5% milk in TBS-T and probed with appropriate primary and secondary antibodies. Protein bands were visualized with enhanced chemiluminescent (ECL) substrate (Lumigen, Southfield, MI, Cat # TMA-100) on a ChemiDoc™ MP Imaging system (Biorad, Hercules, CA, Cat # 12003154).
Co-immunoprecipitation
HEK293T cells were transfected with GLUT4 expression plasmids or RIG-I/MDA5 mutant plasmids using the TransIT-X2 system as described above. Total cell lysates were prepared from transfected cells in lysis buffer containing [ 2% n-Dodecyl-β-D-Maltopyranoside (DDM), 25mM HEPES, 150mM NaCl and 1X protease inhibitors] by gentle scrapping and pipetting before transferring to a 1.5ml centrifuge tube. Lysate was rotated at 4oC for 2h and then centrifuged at 21,000 x g for 10min. Supernatant was removed and again rotated overnight with 20µl of pre-washed (25mM HEPES, 150mM NaCl) anti-FLAG magnetic beads at 4oC. Co-immunoprecipitation was then preformed according to manufacturer’s instructions (Sigma Aldrich, St. Louis, MO, Cat # M8823). IP elution was mixed with 4X SDS loading buffer. Samples were not boiled to maintain GLUT4 transmembrane conformation. For harder to transfect cells such as C2C12, GLUT4 overexpression was carried out in cell suspension. Briefly, cells were dislodged by trypsin digestion and pelleted by brief centrifugation at 800 x g for 5min. The cell pellet was then resuspended in transfection mix (DNA + TransIT-X2 + 1X Opti MEM) prepared as described above for 7–10 min with intermittent agitation at 37oC. Prewarmed DMEM was then added and plated for further culture. FLAG immunoprecipitation then proceeded as above.
Myc immunoprecipitation was performed using 3T3-L1 adipocytes stably expressing the Myc-GLUT4-GFP reporter. Total cell lysates were prepared in lysis buffer (50mM Tris HCl, 150mM NaCl, 1mM EDTA and 1% TRITON X-100) by gentle scrapping and pipetting before transferring to a 1.5ml centrifuge tube. To maintain GLUT4 conformation, lysate was kept on ice for 30min with intermittent vortexing every 5min. Lysate was then centrifuged at 10,000 x g for 10min. Supernatant was removed and mixed with 20µl of pre-washed (25mM Tris, 150mM NaCl and 0.05% Tween-20) Pierce Anti-Myc Magnetic Beads and rotated overnight at 4oC. Co-immunoprecipitation was then preformed according to manufacturer’s instructions (Thermo Fisher Scientific, Waltham, MA, Cat # 88842). IP elution was mixed with 4X SDS loading buffer. Samples were not boiled to maintain GLUT4 transmembrane conformation.
Endogenous MAVS co-immunoprecipitation
For detecting endogenous RIG-I: MAVS interactions, ~ 6 x 106 cells were lysed in 500µl of IP lysis buffer (50mM Tris HCl, 150mM NaCl, 1mM EDTA and 1% TRITON X-100). Cell debris was clarified by centrifuging at 7000 x g for 15min at 4oC and supernatant then mixed with rabbit anti-MAVS (Cell Signaling Technology, Danvers, MA, Cat # 24930) diluted 1:60 (~ 550ng antibody). Lysate and antibody were rotated overnight at 4oC. The antigen sample/antibody mixture was then added to 1.5ml microcentrifuge tube containing 20µl (0.20mg) of prewashed Pierce Protein A/G Magnetic beads (Thermo Fisher Scientific, Waltham, MA, Cat # 88802) and incubated at RT for 2h with mixing. Beads were collected with a magnetic stand and washed thoroughly 3X (1x TBS) as detailed in user manual. Target complexes were eluted in 100µl SDS-PAGE sample lysis buffer on a rotator for 20min. Beads were collected with a magnetic stand and remaining elution was boiled for 10min. To promote RIG-I: MAVS interactions, 3p-hpRNA was transfected into cells as described in “Ligand treatments” section for indicated times detailed in figure legends; reconstitution water diluted to the same ratio in transfection mixture was used for mock groups.
Endogenous MAVS aggregation
To detect MAVS oligomers, C2C12 cells were seeded in 6-well plates at a density of ~ 1 x 106 per well in duplicate and treated as denoted in figure legends. After treatment, cells were washed with 500 µL cold PBS before being lysed in 2% DDM lysis buffer (n-Dodecyl-β-D-Maltopyranoside, 25mM HEPES, 150mM NaCl and 1X protease inhibitors] by gentle scrapping and pipetting. Lysate was then rotated at 4oC for 2h and centrifuged at 21,000 x g for 10min to remove cell debris. Next, supernatants were transferred and mixed with 20 µL Native Sample Buffer (Bio-Rad #1610738) that was then loaded into a 4–16% NativePAGE™ (Invitrogen™ #BN1004BOX). The samples were then run in an Invitrogen NuPage® Novex® Gel System at 120 V for 1 h, 180 V for 30 min, 240 V for 30 min and then 300 V for 1 h in sequence (to maintain at 9 mA electricity). Gels were then transferred to nitrocellulose membranes using Mini Trans-Blot® Cell transfer systems (Biorad, Hercules, CA, Cat #1703930). Membranes were then blocked with 5% milk in TBS-T, probed with appropriate primary overnight and proteins detected with HRP-conjugated secondary antibodies. Protein bands were visualized with enhanced chemiluminescent (ECL) substrate (Lumigen, Southfield, MI, Cat # TMA-100) on a ChemiDoc™ MP Imaging system (Biorad, Hercules, CA, Cat # 12003154).
Plasma membrane fractionation
The plasma membrane fraction was isolated from GLUT4+ C2C12 myocytes and 3T3-L1 adipocytes as previously described, with modifications according to experimental goals 77, 80. Cells were grown on 6-well plates to ~ 2–3 x 106 and serum starved for 3h before 20min of insulin treatment or transfected with 3p-hpRNA for 6h to stimulate innate immune receptors. Culture medium was removed, cells washed with cold 1X PBS twice and then lysed in 180µl Buffer A (50mM Tris HCl, 0.5mM DTT, 0.1% NP-40, 1X protease inhibitors). Cells were scrapped off plates and transferred to a 1.5ml centrifuge tube on wet ice before homogenized with a handheld electric pestle for 15sec (Kimble Chase LLC, United States). The homogenized lysate was passed 3X through a 23G needle attached to a 1mL syringe to shear DNA/nucleus and liberate intracellular proteins (small aliquot removed for total cell lysate and subsequently lysed in RIPA). Current and subsequent centrifugation steps were all done at 4oC to maintain plasma membrane protein structures and prevent denaturing. The remainder of the sheared lysate was centrifuged at 200 x g for 1 min and supernatant transferred to a new pre-chilled tube (Supernatant 1); cell pellet was again homogenized in 90µl Buffer A (0.1% NP-40) until fully dispersed followed by centrifugation for 1 min at 200 x g. Supernatant 1 was combined with Supernatant 2 and the pellet material was discarded, which contained only cell debris. The plasma membrane proteins in supernatant 1 + 2 were centrifuged for 10min at 750 x g to isolate the post-plasma membrane (PPM) fraction and supernatant was then stored on ice for later processing. Again, the cell pellet was suspended in 90µl Buffer A (0.1% NP-40) by vigorously vortexing for 10 sec and centrifuged for 10 min at 750 x g. Cell pellet was resuspended with 45µl Buffer A containing 1% NP-40 (50mM Tris HCl, 0.5mM DTT, 1% NP-40) and incubated on ice for 1h with occasional mixing to solubilize membrane proteins. After 1h, proteins were centrifuged for 20min at 12,000 x g and supernatant containing pure plasma membrane (PM) proteins was removed and mixed with SDS-PAGE sample lysis buffer. The backup PPM fraction was incubated on ice for 1h with occasional mixing and centrifuged at 12,000 x g and mixed with SDS-PAGE sample lysis buffer. Total cell lysate (TCL) was incubated with RIPA for 60 min and centrifuged at the same conditions with PM and PPM fractions. Caveolin was used as loading control for the PM fraction and GAPDH for TCL.
Subcellular fractionation
Subcellular fractionation by differential centrifugation was performed as previously described 38, with slight modifications. Briefly, myotubes were washed 1X with PBS before adding sucrose homogenization buffer (250mM sucrose, 50mM Tris-HCl, 5mM MgCl2 and 1X protease inhibitor cocktail) and scrapping cells off plates. Cellular fractions were kept on ice or 4oC for the remaining steps. Cells were broken up with a Dounce homogenizer for 5min (or until cells were observed to be > 90% broken) and then the nuclei/unbroken cells were vortexed for 15sec and centrifuged at 800 x g for 15min. The pellet discarded and the supernatant was again centrifuged at 800 x g for 10min. The supernatant was retained and centrifuged at 11,000 x g for 10min to separate the cytosol fraction (supernatant) and crude mitochondria (pellet). Crude mitochondria pellets were resuspended in 200µl sucrose buffer, centrifuged again at 11,000 x g for 10min, and pellets resuspended in lysis buffer (50mM Tris-HCl, 1mM EDTA, 0.5% Triton-X100 and 1X protease inhibitor cocktail). Mitochondrial lysis was sonicated on ice 3X in 5 sec increments with 30sec pauses. The lysis buffer was labeled “mitochondrial fraction’ and ready for SDS-PAGE. The cytosol fraction (containing cytosol and microsomes) were centrifuged at 100,000 x g for 1hr in an ultra-centrifuge. The cytosol in the supernatant was concentrated with 100% ice-cold acetone in -20oC for 1hr while the microsome pellet was discarded. The cytosolic fraction was again centrifuged at 12,000 x for 5min, and the pellet containing precipitated cytosolic proteins was resuspended in lysis buffer containing SDS, 10% β-mercaptoethanol and 8M urea.
Immunofluorescence microscopy
Cells were cultured as experiment requested and treated as described in figure legends. For most experiments, C2C12 or 3T3-L1 adipocytes were differentiated in 8-well chambered microscopy slides precoated with IbidiTreat (Ibidi, Gräfelfing, Germany, Cat # 80826) to facilitate attachment for extended periods of time. For plasma membrane staining, medium was removed, and cells washed 2X with cold 1X PBS for 5 min per wash. Cells were subsequently stained for 15 min at 37oC with CellBrite™ Fix 555 (Cat # 300088-T) diluted 1:1000 in PBS. Experiments that did not require plasma membrane staining were immediately washed and fixed as described below. Cells were washed 1X with PBS for 5 min and fixed for 15 min with 4% paraformaldehyde in PBS at RT. Cells were then washed twice with cold 1X PBS for 5 min per wash and permeabilized with 0.1% TRITON X-100 for 15 min at RT. Permeabilization solution was removed and cells were washed 1X with PBS and blocked for 1h at RT in blocking buffer (1% BSA, 10% goat serum, 0.3M glycine, 0.25% TRITON X-100). Primary antibodies (anti-RIG-I, 1:100; anti-GFP, 1:100; anti-FLAG, 1:200) were diluted in antibody buffer (1% BSA, 0.3M glycine, 0.25% TRITON-X100) and incubated with cells overnight at 4oC in the dark. Cells were washed 3X with cold 1X PBS for 5 min per wash. A mix containing a variety of Alexa Flour™ -conjugated secondary antibodies diluted 1:200 (see “Reagent and antibodies” section for specific species reactivity and fluorophore) in antibody buffer was subsequently added to cells for 1h in the dark at RT. Cells were then washed 2X with cold 1X PBS and incubated with DAPI counterstain (Thermo Fisher Scientific, Cat # D1306) diluted 1:1000 in antibody buffer for 10 min in the dark at RT. DAPI was removed and cells were washed once for 5 min in 1X PBS. For chambered slides, cells were covered with 200µL of new 1X PBS; cover slips were mounted onto microscope slides if cells were processed on glass slides. Slides were imaged using a Zeiss 880 laser scanning microscope with a x63.0 oil objective lens.
Experiments involving detection of endogenous GLUT4 were processed as described in the literature and in consultation with Integral Molecular (Philadelphia, PA) 33, 72. C2C12 cells were differentiated in Ibidi chambered microscopy slides and treated as detailed in figure legends. Cells were washed three times with cold 1X PBS for 5 min per wash on ice. Cells were subsequently blocked with 10% goat serum in PBS for 30 min in ice before the addition of exofacial GLUT4 antibody at 10µg/mL in 10% goat serum (Clone LM048, Integral Molecular). Primary antibodies were incubated for 2h at 4oC in the dark and then washed thrice with cold 1X PBS. Cells were subsequently fixed with 4% paraformaldehyde in PBS for 5 min on ice and then moved to RT for an additional 20 min. After fixation, cells were washed twice with 1X PBS and incubated with 50mM glycine in PBS for 5 min at RT. Cells were washed twice with cold 1X PBS before the addition of mixture containing secondary antibody diluted 1:200 (goat anti-human Alexa Fluor™ 488 Cat # A11013) and Hoechst 33342 (Thermo Fisher Scientific, Cat # 62249) diluted 1:10,000 in 10% goat serum for 1h at RT. Secondary antibody was removed and cells were washed three times with 1X PBS. For co-staining of extracellular GLUT4 and intracellular RIG-I, cells were first processed and stained with primary anti-GLUT4 as described above. Then, cells were fixed as before and permeabilized with 10% goat serum containing 0.2% saponin for 15 min at RT. Cells were subsequently blocked intracellularly with 10% goat serum for 30 min on ice and then incubated with primary antibody diluted to 5µg/mL in 1% BSA, 0.3M glycine and 0.1% saponin. Intracellular staining was allowed to proceed overnight at 4oC in the dark. After overnight incubation, cells were washed thrice with 1X PBS on ice and secondary antibody staining proceeded as before. At the secondary antibody step, phalloidin was used to mark the plasma membrane.
All microscopic images were first processed in the Zen 2.3 microscopy software (Zeiss Group, Jena, Germany) before further analysis in ImageJ (NIH). The same microscope instrument settings were used for all samples, and all images were analyzed using the same settings. A vector (white arrow) was drawn from the edge of the nucleus (DAPI) to the plasma membrane (CellBrite 555); the midpoint of the indexed arrow was then considered the perinuclear and peri-plasma membrane regions, respectively. The relocalization of RIG-I intensity (AF488) was calculated by the ratio of fluorescence in the perinuclear to the peri-plasma membrane regions. Fluorescence intensities of all three markers were then traced along the vector and distance (in µm) of RIG-I to the plasma membrane was calculated. Colocalization of membrane and RIG-I fluorescence was measured on a per-cell basis and Pearson’s coefficient calculated using the JACoP Plugin in ImageJ. Mean fluorescence intensities (MFI) for surface GLUT4 were calculated on a per-image basis and identical settings used for each image. In brief, images were adjusted using “Threshold” setting to fix a cutoff value as to highlight regions with fluorescence above the background intensity. Final pmGLUT4 MFI = MFI of cell-background MFI. Because GLUT4 antibody recognizes only exofacial GLUT4 and the cells were not permeabilized or fixed before staining, detected fluorescence was considered on the plasma membrane.
For immunostaining of iPSC-derived cardiomyocytes (iPSC-CMs), cells were fixed with 4% paraformaldehyde in PBS for 12 min at RT, washed with PBS and permeabilized with 0.01% Triton X-100 for 20 min at RT. iPSC-CMs were then blocked with 10% goat serum in PBS for 30 min at RT and stained for cardiac troponin T (cTnT) and nuclei with DAPI (SouthernBiotech, Cat #0100 − 20). Confocal fluorescent microscope (Leica; Multiphoton Microscope TCS SP8 MP) was used to image immunostained cardiomyocytes.
RNA sequencing analysis from patient datasets
RNA sequencing datasets were obtained from PRJNA491748 (CIM study) and GSE143323 (DM study). For transcriptomic analysis of GSE143323, differentially expressed genes (DEGs) were re-identified using DEseq2. FDR/Benjamini-Hochberg method was used to adjust p-values and DEGs with the cutoff of FDR < 0.05 and Log2FC > 0.5 and used to generate volcano plots. Pathway analysis using QIAGEN Ingenuity Pathway Analysis (IPA) included 2361 down-regulated genes and 4417 up-regulated genes and all significant (p < 0.05) pathways with Z-scores were plotted. Selected pathways that were either activated (red) or suppressed (blue) were labeled. Pearson r correlations were conducted on FPKMs values.
Graphing and statistics
All statistical analysis was performed using Graphpad Prism9 software. All experiments were performed in either duplicate or triplicate. Survival curves were analyzed using a log-rank (Mantel–Cox) test. For animal studies, an unpaired, two-tailed non-parametric Mann-Whitney U test was applied to statistical analysis. For comparison in which multiple variables were tested with multiple time points, two-way ANOVA analysis was performed. For comparison of two data points in vitro or in vivo, the student’s two-tailed unpaired Student’s t test was applied. The sample sizes (biological replicates), specific statistical tests used, and the main effects of our statistical analyses for each experiment were detailed in each figure legend. Asterisk coding is as follows: * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001. Data with error bars depict the average with the SEM.