AUF-1 knockdown in mice undermines gut microbial butyrate-driven hypocholesterolemia through AUF-1–Dicer-1–mir-122 hierarchy

Introduction and objective Cholesterol homeostasis is a culmination of cellular synthesis, efflux, and catabolism to important physiological entities where short chain fatty acid, butyrate embodied as a key player. This discourse probes the mechanistic molecular details of butyrate action in maintaining host-cholesterol balance. Methods Hepatic mir-122 being the most indispensable regulator of cholesterol metabolic enzymes, we studied upstream players of mir-122 biogenesis in the presence and absence of butyrate in Huh7 cells and mice model. We synthesized unique self-transfecting GMO (guanidinium-morpholino-oligo) linked PMO (Phosphorodiamidate-Morpholino Oligo)-based antisense cell-penetrating reagent to selectively knock down the key player in butyrate mediated cholesterol regulation. Results We showed that butyrate treatment caused upregulation of RNA-binding protein, AUF1 resulting in RNase-III nuclease, Dicer1 instability, and significant diminution of mir-122. We proved the importance of AUF1 and sequential downstream players in AUF1-knock-down mice. Injection of GMO-PMO of AUF1 in mouse caused near absence of AUF1 coupled with increased Dicer1 and mir-122, and reduced serum cholesterol regardless of butyrate treatment indicating that butyrate acts through AUF1. Conclusion The roster of intracellular players was as follows: AUF1-Dicer1-mir-122 for triggering butyrate driven hypocholesterolemia. To our knowledge this is the first report linking AUF-1 with cholesterol biogenesis.


In silico analysis of global microarray data
Raw microarray expression data was obtained in a CEL file format using GSEOquery package (DOI: 10.18129/B9.bioc.GEOquery) from GSE45220 (human) and GSE4410 (mice) datasets. The GEOquery R package parses GEO data into R data structures that can be used by other R packages. Normalization and background correction of Gene Expression Measurements was performed using oligo package's Robust Multichip Average (RMA) algorithm that attempts to remove local biases across samples in order to enable relevant differential expression testing. Variations are scanned across the samples using principal component analysis and Hieracrchial clustering method. Following which, differential gene analysis was done using Limma package. The limma (Linear Models for Microarray Analysis) R package has evolved as the most widely used statistical tests for identifying differentially expressed genes using Limma based adjusted P-value. P-value of <0.05 are selected. Results are annotated using library hugene10sttranscriptcluster.db for GSE45220, and huex10sttranscriptcluster.db for GSE4410. In case of genes with two or more probes aligned/mapped, only the most significant ones were chosen and the list of upregulated and downregulated genes are prepared accordingly. Genes with the smallest adjusted P-value and highest logFC values are the most significant. Amongst them genes which have a possible role in cholesterol metabolism were selected using KEGG pathway, BioGPS and Reactome databases and were sorted accordingly. All the cholesterol genes showed a downregulated expression. The data were represented as Volcano plot.

Isolation for Murine Hepatocytes:
Murine hepatocytes were isolated via Percoll method (1) with slight modifications. Briefly, euthanized C57BL/6 mice were placed on top of tissue paper pillow and incision was done to open the abdomen properly. A small incision was done to insert cannula into the portal vein and liver was perfused with 10 ml of perfusion buffer (25mM HEPES and 0.5M EDTA in HBSS without Ca 2+ and Mg 2+ ). Digestion buffer (25mM HEPES and 25 µg/ml Collagenase IV in HBSS with Ca 2+ and Mg 2+ ) was passed through the vena cava. Liver was carefully removed into the petridish containing digestion buffer and dissociated carefully. Cells were filtered through 100µm cell strainer (BD) and centrifuged for 5 minute at 50g. Supernatant was discarded and pellet was collected, to which 10ml of plating media (DMEM+10%FBS) was added. Subsequently, 90% of Percoll was laid over this solution and centrifuged for 10 minutes at 300g at 4°C. Supernatant was discarded and cells were washed and suspended in 10ml of DMEM.

Transmission Electron Microscopy
Liver tissue was fixed with 3% glutaraldehyde in 0.1M sodium cacodylate buffer.
Subsequently, a secondary fixation was conducted with 1% Osmium tetroxide, followed by dehydration with ascending grades of acetone, and finally embedded in Agar 100 resin and polymerization at 60°C. The ultrathin sections (40-50 nm) of the tissue were obtained using a Leica Ultracut UCT ultramicrotome (Leica Microsystems, Germany), picked up on nickel grids, and dual-stained with 2% aqueous uranyl acetate and 0.2% lead citrate. The sections were visualized under a FEI Tecnai 12 Biotwin transmission electron microscope (FEI, Hillsboro, OR, USA) at an accelerating voltage of 100 kV (2).

Food consumption
The daily consumption of food in mice was recorded every day by weighing the food given (food placed in the food receptacle at time zero) and food remaining (food left in the food receptacle 24 hours later). The cumulative food intake by each group (5 mice / group) per day was determined.

Collection of blood, preparation of serum samples, biochemical analysis, estimation of cholesterol and lipoproteins and hepatic enzymes
Blood collected from tail vein was allowed to stand for 3 h at room temperature and then serum was prepared by centrifugation at 1800 rpm analyzed for serum cholesterol by Assay Kit. Levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) in serum were measured in using kit from Transasia Biomedicals (India).

Transfection
Cells were plated in 6-well culture plates at the density of 1×10 6 cells/well and cultured in 2 ml serum-free medium for 24 h to 80% confluence. Transfection was performed with

Fluorescence microscopy
Cells grown on glass cover slips were transfected with either pEGFP-AUF1 p40 plasmid (4) or pEGFP-Dicer-1-3′UTR or p-EGFP (5) for 24 h. After washing with PBS, the cells were treated with 1 µg/ml Hoechst 33342 for 5 min at room temperature and washed again with PBS three times. Fluorescence images were captured with Carl Zeiss microscope equipped with a CCD camera controlled with ZEN software (Carl Zeiss, Gottingen, Germany).

Tissue homogenisation, preparation of RNA and Protein
Liver samples were dissected into small pieces and were resuspended either in RIPA Lysis buffer (20mM Tris-HCl pH 7.5, 150 mM NaCl, 1mM EDTA, 1mM EGTA, 1% NP-40, 1% Sodium deoxycholate, 2.5 mM Sodium Pyrophosphate, 1mM β-glycophosphate, 1mM Na 3 VO 4 , 1 µg/ml leupeptin with 1mM of PMSF immediately before use) for protein isolation or in Trizol (Invitrogen, US) for RNA isolation. The tissue was homogenized using a micropestle and centrifuged at 13,000 g for 15 min at 4°C. The clear supernatant was collected and either stored as protein lysate in -80°C or further processed to isolate RNA using the standard protocol (3).

RNA extraction and reverse transcription
Cells were cultured in 24-well plates to 80% confluence. Total RNA from cells or tissue was extracted with Trizol according to the protocol recommended by the manufacturer. The concentration of the extracted RNA was analyzed by Nanodrop spectrophotometer (Thermo) and RNA was stored at -80° C. cDNA was prepared from total RNA by reverse specific primers using Super Reverse Transcriptase MuLV Kit. The primers for the reverse transcription are listed in Table 1. U6 and GAPDH were normalized for the expressions of miRNAs and other genes of interest respectively. The total reaction volume for reverse transcription was 20 μl in which 1 μM of reverse primer, 5 ng of RNA template, 1 μl dNTP mix, 12 μl of DEPC treated water, 4 μl of 5X first strand buffer, 1 μl of 0.1 M DTT, 1 μl of RNase inhibitor and 1 μl Super RT MuLV. Reverse transcription was carried out for 65°C for 5 minutes, followed by incubation at 55°C for 1 hour and then heat inactivating the reaction at 70°C for 15 minutes (5).

Quantitative real-time PCR
The miRNAs and mRNA levels were quantified with Applied Biosystems TM StepOne TM Real Time PCR System with RT 2 SYBR ® Green qPCR Mastermix following the manufacturer's instructions. Each 20 μl qPCR reaction contained an amount of cDNA equivalent to 5 ng of total RNA, 10 μl of RT 2 SYBR ® Green qPCR Mastermix, 1 μM of the forward and reverse primer (each) and nuclease free water (6). Real-time PCR was performed with the following conditions: 95°C for 10 min, 40 cycles of 95°C for 30 sec, 60°C for 1 min and 72°C for 1 min PCR product and was normalized either with GAPDH Ct value or U6 (for microRNA) as described previously (5).
The stem loop primer stock was overlaid with 100µl molecular biology grade mineral oil (Sigma). The mixture was heated to 95°C and were kept at 75°C, 68°C, 65°C, 62°C and 60C for an hour each respectively. Thereafter, working stock of 10µM was prepared and stored in -20°C till further use (7).

Western blot
Liver tissue protein or cell lysate were extracted in RIPA Lysis buffer.

Synthesis of Fmoc protected thiourea morpholino active monomer
For the synthesis of GMO (Guanidinium Morpholino Oligonucleotides, GMO) part of GMO-PMO chimera, the Fmoc protected thiourea MMTr-morpholino active monomers of C, A and G nucleosides were required. They were synthesized using the modified protocol as described earlier (8).

Synthesis of Trityl protected chlorophosphoramidate morpholino active monomer
For the synthesis of PMO we have used chlorophosphoramidate monomers of A, T, G and C which was synthesized as per earlier report (9).

Functionalization of solid support with linker and loading monomer
Prior to solid phase synthesis of GMO-PMO sequences, polystyrene solid support was functionalized with aminocaproic acid linker and loading monomer as per earlier report (10).

Solid phase synthesis of GMO-PMO (Fig S9 A)
After the successful incorporation of linker and loading monomer on polystyrene solid support, coupling for GMO synthesis was initiated. For GMO synthesis 5 equivalent of Fmoc protected thiourea, 5 equivalents of HgCl 2 and 5 equivalent of NEM were added in NMP solvent. This step was repeated for another two times in 2 hrs interval. Total coupling time per GMO unit was 6 hrs. Excess reagents were washed with 20 % Thiophenol-NMP and NMP. Unreacted amine was capped with (1:1) mixture of 10 % Ac 2 O-NMP and 10 % DIPEA-NMP. MMTr group was deprotected using the deblocking cocktail (CYPTFA) (10, 11). The synthetic cycle (washing, coupling, capping and deblocking) was repeated for another three GMO monomers (see the structures above) to get the GMO unit. The GMO pentamer was further reacted with chlorophosphoramidate monomer (see the structure above). The morpholino part was synthesized as per our reported method (11). Full length GMO-PMO was cleaved from solid support using 33 % aqueous ammonia at 55ºC for 16 hrs and purified by acetone precipitation. Purity of the synthesized GMO-PMOs were checked in HPLC (Fig S9B and C)      sets, data is represented as mean SE. *** represents p<0.001, ** represents p<0.01, * represents p<0.05, ns represents not significant.

Figure S8
Antibiotic treatment increases serum cholesterol The serum cholesterol of normal-mice (A) and Abx-mice (B) on day 7 and day 21.  used in this study was specific to AUF-1 p37 . N=2