An analog of glibenclamide selectively enhances autophagic degradation of misfolded α1-antitrypsin Z

The classical form of α1-antitrypsin deficiency (ATD) is characterized by intracellular accumulation of the misfolded variant α1-antitrypsin Z (ATZ) and severe liver disease in some of the affected individuals. In this study, we investigated the possibility of discovering novel therapeutic agents that would reduce ATZ accumulation by interrogating a C. elegans model of ATD with high-content genome-wide RNAi screening and computational systems pharmacology strategies. The RNAi screening was utilized to identify genes that modify the intracellular accumulation of ATZ and a novel computational pipeline was developed to make high confidence predictions on repurposable drugs. This approach identified glibenclamide (GLB), a sulfonylurea drug that has been used broadly in clinical medicine as an oral hypoglycemic agent. Here we show that GLB promotes autophagic degradation of misfolded ATZ in mammalian cell line models of ATD. Furthermore, an analog of GLB reduces hepatic ATZ accumulation and hepatic fibrosis in a mouse model in vivo without affecting blood glucose or insulin levels. These results provide support for a drug discovery strategy using simple organisms as human disease models combined with genetic and computational screening methods. They also show that GLB and/or at least one of its analogs can be immediately tested to arrest the progression of human ATD liver disease.


Genomic engineering of cell lines.
For targeted disruption of ATG14 in the HTO/Z cell line, ATG14 was targeted at exon 5 with guide RNA 5′-AAGAAGTCATTATGAGCGTC-3' cloned into the MLM3636 gRNA expression vector (gift from Keith Joung, Addgene plasmid #43860) as previously described (1). HTO/Z cells (1 x 10 6 ) were transfected using 500 ng guide RNA plasmid and 2 ug of Cas9 expression plasmid (gift from Jin-Soo Kim, Addgene plasmid #43945) via nucleofection using 100 ul of solution P3 and program CN-114 (Lonza 4D-Nucleofector, X unit). Transfected pools were single cell sorted into 96-well plates using a MoFlo Cell Sorter and expanded for 2-3 weeks prior to harvesting. Targeted deep-sequencing was performed to determine genotype and clone purity using target specific primers forward 5' TCTGAAGGCCTTCTCAAAACCAAGGA 3' and reverse 5' TCTGCGGTGCGTACTGTTTGTTGAA 3'. PCR was used to confirm the presence of the ATZ (forward 5' ACCCGGGTCGAGTAGGCGTGTA 3' and reverse 5' GCTGGCTGTCTGGCTGGTTGA 3') and TET (forward 5'CATTGACGCAAATGGGCGGTAG and reverse 5' GCCAATACAATGTAGGCTGCTCTAC 3') expression plasmids for each clone. All PCR amplifications were performed with MyTaq Red Mix (Bioline), according to the manufacturer protocol. Cell lines were tested for mycoplasma contamination.
SiRNAs. ON-TARGETplus SMART pool siRNA for MRP1 (Dharmacon: L-007308-00) and MRP3 (Dharmacon: L-007312-00) were used for the investigation of ABC transporters. ON-TARGETplus Non-targeting Pool was used as a control (Dharmacon: D-001810-01-20). The HTOZ cells were transfected at a final concentration of 30nM with Lipofectamine 2000 (Thermo Fisher Scientific). The cells were incubated for 48 hours in the presence or absence of GLB (40µM) after the transfection. Samples were subjected to real-time RT-PCR for MRP1 and MRP3 mRNA expression and then immunoblot analysis for ATZ and β-actin. proteasomal inhibitors were used, MG132 was used at 30 μM for the last 6 hours of the incubation with GLB or control. Cells that were incubated with MG132 alone served as control to validate that the proteasome was inhibited.
For investigation of LC3 conversion, lysosomal inhibitors (E64D and pepstatin A at 20 μg/ml) were added to the medium for the last 4 hours (or overnight) of the incubation with GLB or control. This has been shown to inhibit the lysosomal degradation of LC3-II and when compared to the LC3-II levels in the absence of lysosomal inhibitors to provide a reflection of autophagic flux. Other lysosomal inhibitors include chloroquine (CLQ) at 50µM, ammonium chloride at 10 mM, used in overnight incubation time periods. Bafilomycin was used at a concentration of 100 nM in overnight incubation trials.
Insulin Secretion Assay. Pancreatic β cells Min6 were maintained in DMEM high glucose with the supplemental of 10% FBS, 1% penicillin/streptomycin and 5 μl/L 2-mercaptoethanol. Insulin secretion assays were performed using insulin ELISA kits from Crystal Chem INC (Downers Grove, IL). Briefly, cells were plated in 24 well plates 48-hour before the assay. The cells were treated with increasing doses of GLB or the analogs ranging from 1 nM to 10 μM. On the day of experiments cells were washed twice and incubated with Krebs-ringer bicarbonate buffer (KRBB) containing 2.5 mM glucose and 0.1% BSA for 1 hour. Buffer was removed from the cells and fresh KRBB containing chemicals (GLB or analogs) or 16 mM glucose (used as positive control) were added to the cells and incubated for 1 hour. The supernatant was collected right after the incubation and insulin concentration was assessed according to manufacturer's instructions.
For immunoblot analysis of proteins in liver, the liver was snap frozen in liquid nitrogen and stored at −80°C. Liver was homogenized in 50 mM Tris-HCl pH 8.0, 150mM NaCl, 2 mM KCl, 2 mM MgCl2, 0.5% Triton X-100, 0.5% deoxycholic acid containing 0.1 mM phenylmethylsulfonic acid and Complete protease inhibitor cocktail from Roche. Total protein concentration was measured by BCA assay (Pierce) and followed by western blotting as described previously. Real Time Quantitative PCR. Total RNA was extracted from PiZ mouse livers or HTO/Z cells with TRIZOL (Life Technologies, Grand Island, NY). First strand cDNA (RT reaction) was synthesized from 2 µg of RNA using high capacity RNA-to-cDNA kit (Applied Biosystems, Foster City, CA) according to the manufacturer's directions. A negative control was performed without enzyme (NRT reaction). RT and NRT reactions were also performed on 2 µg of commercially prepared liver RNA (Ambion, Austin, TX) to serve as the calibrator for the real time QPCRs. Each experimental sample was normalized to a nontransgenic control (fold change). For PCR, duplicate aliquots of the RT reaction and 1 aliquot of the NRT reaction served as templates for the target genes and the control gene β-glucuronidase (GusB). The probes and primers were obtained from Applied Biosystems (SERPINA1 assay ID: Hs01097800-m1; mouse GusB assay ID: Mm00446953-m1; human GusB assay ID: Hs9999908-m1; human MRP1 assay ID: Hs01561483_ml; human MRP3 assay ID: Hs00978452_ml). Real time reactions were run on an ABI7300 using the following cycling parameters: 95°C for 12 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. Differential gene expression was calculated by the ΔΔCT calculation (4). The ΔΔCT method controls for potential differences in efficiency of the RT, as well as the PCR, whereas calculations based on standard curves do not (ΔΔCT = ΔCT(exp) -ΔCT (ctr) when ΔCT = ΔCT(gene) -ΔCT(norm)).

Radio immunoprecipitation and SDS
Analysis of experiments in PiZ mouse. The outcome was determined by evaluating 1) the hepatic ATZ load by immunoblotting with anti-AT, immunostaining with PAS/D; 2) hepatic fibrosis using Sirius Red staining; 3) autophagy by immunoblotting with anti-LC3 and anti-p62, and imaging for GFP+ autophagosomes; 4) monitoring blood glucose levels by glucometer and blood insulin levels by ELISA Wide Range Assay (CrystalChem). For immunostaining, liver samples were fixed in 10% Formalin, followed by staining with PAS after diastase and Sirius Red using standard techniques (2). Quantitative evaluation of immunostaining and histochemical staining was carried out by a member of the team that was blinded to group allocation.

Statistical Analysis. Sample size was initially chosen based on previous studies with other
drugs that had a statistically significant effect and subsequently was based on results of the initial experiments. No animals, samples, or data points were excluded from the reported analysis once obtained. Student's t-test was used for most comparisons but the Welch modified t-test was used to compare experimental groups that were not paired and did not assume equal variances. Kinetic and dose-response curves were analyzed by two-way ANOVA with the Bonferroni post-test using the Prism software application. All analyses were considered statistically significant at P<0.05.    The relative effect of GLB on ATZ levels in the absence or presence of lysosomal enzyme inhibitors was determined by densitometric scanning. The results are reported here as % change for the effect of GLB. In the absence of lysosomal inhibitor is described as the control (bar labelled CTR, n=16). This is compared to % change for GLB in samples also treated with NH4Cl2 (n=4), chloroquine (n=4) or bafilomycin (n=8). *Significant difference compared to CTR (p<0.05, unpaired Student T-test).  GAPDH was used as the loading control. ATG5 and LC3 levels were analyzed to verify the decrease in ATG5 levels and in LC3-II to LC3-I ratio.

Animal Committee Approvals. All animal experiments were approved by the IACUC at
(2) HTOZATG5KD cells were incubated for 48 hours in the absence or presence of GLB and bafilomycin 100 nM was added to separate aliquots for the last 15 hours of the incubation period with analysis for ATZ and β-actin levels as the read-out. . p70 S6K is marked as S6K, phospho-p70 S6K is marked as pS6K; phospho-4E-BP1 is marked as p4E-BP1 on the figure. (2) HTO/Z cells were incubated with or without GLB for 48 hrs. and then examined by immunoblotting with antibodies to the UPR proteins, including BiP, GRP94 and calnexin together with tubulin as loading control.   Immunoblot analysis for p62. The difference was significant (p=0.0363). Statistical analysis used unpaired two-tailed t-test.