The DNA glycosylase NEIL2 is protective during SARS-CoV-2 infection – Nature.com

Ethics statement

Human Study: The lung specimens from the COVID-19 positive human subjects were collected using autopsy (study was IRB Exempt). All donations to this trial were obtained after telephone consent followed by written email confirmation with next of kin/power of attorney per California state law (no in-person visitation could be allowed into the COVID-19 ICU during the pandemic). The detailed patient characteristics were published elsewhere (PMID: 34127431). For normal lung tissues, lung biopsies were obtained after surgical resection of lungs by cardiothoracic surgeons as before [https://elifesciences.org/articles/66417]. Deidentified lung tissues obtained during surgical resection, which were deemed excess by clinical pathologists, were collected using an approved human research protocol (IRB no. 101590). Blood samples were obtained from UTMB Biorepository of research subjects with a laboratory diagnosis of COVID-19 that consented to participate in the Clinical Characterization Protocol for Severe Emerging Infections (UNMC IRB no. 146-20-FB/UTMB IRB no. 20-0066). The normal healthy subjects blood cell pellets were obtained under UTMB IRB no. 14-0131 and 20-0097.

Animal (Hamster) study: Lung samples from 8-week-old male Syrian hamsters were generated from experiments conducted exactly as in previously published studies (PMID: 32540903). Animal studies were approved and performed in accordance with Scripps Research IACUC Protocol no. 20-0003 and UTMB IACUC Protocol no. 2005060.

Publicly available COVID-19 gene expression databases were downloaded from the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus website (GEO)69,70,71. If the dataset was not normalized, RMA (Robust Multichip Average)72,73 was used for microarrays and TPM (Transcripts Per Millions)74,75 was used for RNASeq data for normalization. We used log2 (TPM+1) to compute the final log-reduced expression values for RNASeq data. Accession numbers for these crowd-sourced datasets are provided in the figures and manuscript. Single Cell RNASeq data from GSE145926 was downloaded from GEO in the HDF5 Feature Barcode Matrix Format. The filtered barcode data matrix was processed using Seurat v3 R package76. Pseudo bulk analysis of GSE145926 dataset was performed by adding counts from the different cell subtypes and normalized using log2 (CPM+1). All of the above datasets were processed using the Hegemon data analysis framework77,78,79.

Time (duration in hospital) and status (whether the patient is discharged from hospital) were derived from the hospital-free days post 45-day follow-up from COVID-19 patients (n=100, GSE157103). All non-COVID-19 patients (n=26, GSE157103) were excluded from the analysis. KaplanMeier (KM) analysis is performed using lifelines python package version 0.14.6. All KM analyses use the StepMiner threshold+0.5 noise margin as the threshold to separate the patients into high and low groups.

COVID-19 samples were inactivated by storing in 10 % formalin for 2 days and then were transferred to zinc-formalin solution for another 3 days. The decontaminated tissues were transferred to 70% ethanol and cassettes were prepared for tissue sectioning. The slides containing hamster and human lung tissue sections were de-paraffinized in xylene (Sigma-Aldrich, catalog no. 534056) and rehydrated in graded alcohols to water. For NEIL2 antigen retrieval, slides were immersed in Tris-EDTA buffer (pH 9.0) and boiled for 10min at 100C inside a pressure cooker. Endogenous peroxidase activity was blocked by incubation with 3% H2O2 for 10min. To block non-specific protein binding 2.5% goat serum (Vector Laboratories, catalog no. MP-7401) was added. Tissues were then incubated with rabbit anti-NEIL2 polyclonal antibody (in house generated, 33) for 1.5h at room temperature in a humidified chamber and then rinsed with TBS or PBS 3x, 5min each. Sections were incubated with horse anti-rabbit IgG (Vector Laboratories, catalog no. MP-7401) secondary antibodies for 30min at room temperature and then washed with TBS or PBS 3x, 5min each; incubated with 3,3-diaminobenzidine tetrahydrochloride (DAB) (Thermo Scientific, catalog no. 34002), counterstained with hematoxylin (Sigma-Aldrich, catalog no. MHS1) for 30s, dehydrated in graded alcohols, cleared in xylene, and cover slipped. Epithelial and stromal components of the lung tissue were identified by staining duplicate slides in parallel with hematoxylin and eosin (Sigma-Aldrich, catalog no. E4009) and visualizing by Leica DM1000 LED (Leica Microsystems, Germany).

IHC images were randomly sampled at different 300300 pixel regions of interest (ROI). The ROIs were analyzed using IHC Profiler80. IHC Profiler uses a spectral deconvolution method of DAB/hematoxylin color spectra by using optimized optical density vectors of the color deconvolution plugin for proper separation of the DAB color spectra. The histogram of the DAB intensity was divided into 4 zones: high positive (060), positive (61120), low positive (121180) and negative (181235). High positive, positive, and low positive percentages were combined to compute the final percentage positive for each ROI. The range of values for the percent positive is compared among different experimental groups.

Lung specimens from COVID-19 positive human subjects were collected using autopsy procedures at the University of California San Diego (the study was IRB Exempt) following guidelines from the Centers for Disease Control and Prevention (CDC) and College of American Pathologists autopsy committee. All donations to this trial were obtained after telephone consent followed by written email confirmation with next of kin/power of attorney per California state law (no in-person visitation could be allowed into the COVID-19 ICU during the pandemic). (https://www.cdc.gov/coronavirus/2019-ncov/hcp/guidance-postmortem-specimens.html and https://documents.cap.org/documents/COVID-Autopsy-Statement-05may2020.pdf). Lung specimens were collected in 10 % Zinc-formalin and stored for 72h before processing for histology as done previously81,82.

Blood cell pellets stored in TRIzol LS Reagent (Invitrogen, catalog no. 10296010) were obtained from the UTMB Biorepository for Severe Emerging Infections from research subjects with a laboratory diagnosis of COVID-19 that consented to participate in the Clinical Characterization Protocol for Severe Emerging Infections (UNMC IRB no. 146-20-FB/UTMB IRB no. 20-0066). Samples were used from subjects categorized as having moderate or severe COVID-19 based on the following criteria: moderate disease if requiring oxygen via nasal cannula, severe disease if requiring oxygen via non-invasive ventilation (e.g., CPAP, BiPAP, High-Flow nasal cannula, venturi mask). The normal healthy subjects blood cell pellets were obtained in TRIzol LS Reagent under UTMB IRB # 14-0131 and 20-0097. Total RNA was isolated as per manufacturers protocol and subjected to real time reverse transcriptase-quantitative Polymerase Chain Reaction.

Total RNA extraction was performed from cells using TRIzol Reagent (Invitrogen, catalog no. 15596026) or TRIzol LS Reagent. Total RNA (up to 2g) was used to synthesize cDNA with a PrimeScriptTM RT Kit with gDNA Eraser (TaKaRa, catalog no. RR047A) and qPCR was carried out using TB Green Premix Ex Taq II (Tli RNase H Plus; TaKaRa, catalog no. RR820A) in Applied Biosystems 7500 Real-Time PCR Systems with thermal cycling conditions of 94C for 5min, (94C for 10s, and 60C for 1min) for 40 cycles, and 60C for 5min. The target mRNA levels were normalized to that of GAPDH or 18S RNA. Primer sequences used in the assay are listed in Supplementary Table1. In each case, DNase-treated RNA samples without reverse transcriptase were amplified to test genomic DNA contamination.

Syrian golden hamsters (Hamster/Golden Syrian Hamster/Male/8 weeks old/Charles River/Strain Code 049) experiments were approved by the Scripps Research Institute Institutional Animal Care and Use Committee/Protocol 20-0003, and were carried out in accordance with recommendations. Lung samples were collected from 8-week-old Golden Syrian hamsters post SARS-CoV-2 infection conducted exactly as in a previously published study38. Briefly, lungs from hamsters challenged with SARS-CoV-2 (1106 PFU) were harvested on day 5 (peak weight loss) and NEIL2 protein and mRNA levels were analyzed by IHC and RT-qPCR, respectively. Syrian golden hamsters (Male/8 weeks old) were infected with SARS-CoV2 as approved by the UTMB IACUC (protocol no. 2005060) and nuclear extract was prepared from the uninfected and infected hamster lungs at 10 days post infection as described before33,39, and DNA was extracted from the same samples for LA-qPCR.

A549 cells stably expressing human angiotensin I converting enzyme 2 (A549-ACE2)83 is maintained in Eagles Minimum Essential Media (EMEM; Gibco, Cat # 11095080), containing 10% fetal bovine serum (FBS), 100units/ml penicillin and 100g/mL streptomycin. A549-ACE2 cells grown in six-well plates at ~70% confluence were transduced with recombinant proteins using Pierce Protein Transfection Reagent according to manufacturers recommendations (Pierce, Thermo Scientific, catalog no. 89850). In brief, Pierce reagent (dissolved in 250L of methanol or chloroform) was evaporated to remove traces of solvent and 2g of rNEIL1, or rNEIL2 protein was added in PBS, vortexed, incubated for 5min at room temperature, then the mixture was supplemented with serum free medium. Mixtures were added directly onto the cell monolayers, incubated for 4h in a 5% CO2 containing incubator at 37C and then one volume of 20% serum-containing medium was added for overnight. Transfection efficiency varied between 68 and 75% as determined in parallel experiments by indirect immunofluorescence assays using anti-NEIL2 or anti-NEIL1 (in house generated84) antibodies. Transduced A549/ACE2 cells were infected with SARS-CoV-2 at MOI 11.87. After incubation for an hour with viral inoculum, cells were washed three times with EMEM. Infected cells were harvested at indicated time points in various lysis buffers, depending on the downstream experiment. Supernatants from infected cells were harvested at 24h post-infection for measuring the infectious virus titers by the TCID50 assay using Vero E6 cells. Briefly, 50L supernatants from infected cells were serially diluted (10-fold) in EMEM supplemented with 2% FBS; 100L of serially diluted samples were added to Vero E6 cells grown in 96-well plates and cultivated at 37C for 3 days followed by observation under a microscope for the status of virus-induced formation of cytopathic effect (CPE) in individual wells. The titers were expressed as log TCID50/mL.

Human bronchial epithelium cell line, BEAS-2B (ATCC CRL-9609) stably expressing NEIL2-FLAG,human gastric adenocarcinoma (AGS, ATCC CRL-1739) and human embryonic kidney cells (HEK29385) were grown at 37C and 5% CO2 in DMEM/F-12 (1:1) containing 10% FBS, 100units/ml penicillin and 100units/ml streptomycin. For all experiments, 5060% confluent cells were used. We routinely tested cell lines for mycoplasma contaminations using the PCR-based Venor GeM Mycoplasma Detection Kit (Sigma, catalog no. MP0025). Control or stable BEAS-2B cells at ~70% confluency were transiently transfected with vector expressing GFP with (SARS-CoV2-5-UTR-eGFP construct, synthesized and cloned by GenScript Inc.) or without (UTR-Less-eGFP construct) UTR (100ng) using Lipofectamine TM 2000 (Invitrogen, catalog no. 11668027), according to the suppliers protocol. To monitor transfection efficiency, a reporter gene construct (0.25g) containing -galactosidase downstream to the SV40 promoter was co-transfected. Cells were allowed to recover for 16h in media with serum and then GFP florescence was measured using an ECHO florescent microscope (ECHO Revolve-R4). Total RNA and DNA were isolated for subsequent qPCR analysis.

The proteins in the nuclear extracts (from Hamster lungs)/whole cell extracts A549-ACE2 cells were separated onto a Bio-Rad 420% gradient Bis-Tris gel, then electro-transferred on a nitrocellulose (0.45m pore size; GE Healthcare) membrane using 1X Bio-Rad transfer buffer. The membranes were blocked with 5% w/v skimmed milk in TBST buffer (1X Tris-Buffered Saline, 0.1% Tween 20) and immunoblotted with appropriate antibodies SARS-CoV-2 spike protein (S1-NTD) (Cell Signaling Technology, catalog no. 56996S), GAPDH (BioBharati Life Sciences, catalog no. AB0060), Histidine (BioBharati Life Sciences, catalog no. AB0010), NEIL233, OGG1 (in-house generated86), NEIL1 and APE1 (in-house generated87), and HDAC2 (Histone deacetylase 2, GeneTex, catalog no. GTX109642). The membranes were extensively washed with 1% TBST followed by incubation with anti-isotype secondary antibody (Cell Signaling Technology, catalog no. 7074) conjugated with horseradish peroxidase in 5% skimmed milk at room temperature. Subsequently, the membranes were further washed three times (10min each) in 1% TBST, developed and imaged using kwikquant image analyzer and image analysis software (ver. 5.2) (Kindle Biosciences). Due to cross reactivity of common secondary antibody with the pre developed membrane, the samples were run in parallel gels in similar conditions, and developed with different antibodies. For all the primary antibodies, 1:1000 dilution was used and for secondary antibody, 1:2000 dilution was used.

RNA-ChIP assays were performed as described earlier39. Briefly, cells were cross-linked in 1% formaldehyde for 10min at room temperature. Then 125mM Glycine was added for 5min at room temperature to stop crosslinking and then samples were centrifuged at 1000g at 4C for 5min to pellet the cells. The cell pellet was re-suspended in sonication buffer, containing 50mM Tris-HCl pH 8.0, 10mM EDTA and 1% SDS with 1X Protease inhibitor cocktail and sonicated to an average DNA size of ~300bp using a sonicator (Qsonica Sonicators). The supernatants were diluted with 15mM Tris-HCl pH 8.0, 1.0mM EDTA, 150mM NaCl, 1% Triton X-100, 0.01% SDS containing protease inhibitors, and incubated with anti-NEIL1, -NEIL2, -FLAG (Millipore, catalog no. F1804) or normal IgG (Santa Cruz, catalog no. sc-2025) antibodies overnight at 4C. Immunocomplexes (ICs) were captured by Protein A/G PLUS agarose beads (Santa Cruz, catalog no. sc-2003), that were then washed sequentially in buffer I (20mM Tris-HCl pH 8.0, 150mM NaCl, 1mM EDTA, 1% Triton-X-100 and 0.1% SDS); buffer II (same as buffer I, except containing 500mM NaCl); buffer III (1% NP-40, 1% sodium deoxycholate, 10mM Tris-HCl pH 8.0, 1mM EDTA), and finally with 1X Tris-EDTA (pH 8.0) buffer at 4C for 5min each. RNase inhibitor (50Uml1, Roche, catalog no. 03335402001) was added to sonication and IP buffers, and 40Uml1 to each wash buffer. The ICs were extracted from the beads with elution buffer (1% SDS and 100mM NaHCO3) and de-crosslinked for 2h at 65C. RNA isolation was carried out in acidic phenolchloroform followed by ethanol precipitation with GlycoBlue (Life Technologies, catalog no. AM9516) as a carrier. Reverse transcription and cDNA preparation was performed using a PrimeScript RT Kit with gDNA Eraser. RNA-ChIP samples were analyzed by qPCR using specific primers (listed in Supplementary Table1) and represented as percentage input after normalization to IgG.

Wild-type recombinant His-tagged -NEIL2, -NEIL2-ZnF mutant (ZnF-NEIL2mut) and -NEIL1 proteins were purified from E. coli using protocol as described earlier64. Briefly, pET22b (Novagen) vector containing C-terminal 6xHis tagged Coding DNA Sequence (CDS) of various proteins was transformed into E. coli BL21(DE3) RIPL Codon-plus cells (Agilent technologies, catalog no. 230280). The log-phase culture (A600=0.40.6) of E. coli was induced with 0.5mM isopropyl-1-thio--D-galactopyranoside (IPTG) and grown at 16C for 16h. After centrifugation, the cell pellets were suspended in a lysis buffer (Buffer A) containing 25mM Tris-HCl, pH 7.5, 500mM NaCl, 10% glycerol, 1mM -mercaptoethanol (-ME), 0.25% Tween 20, 5mM imidazole, 2mM phenylmethylsulfonyl fluoride (PMSF). After sonication, the lysates were spun down at 13,000rpm and the supernatant was loaded onto HisPur Cobalt Superflow Agarose (Thermo Scientific, catalog no. 25228), previously equilibrated with Buffer A, and incubated for 2h at 4C. After washing with Buffer A with a gradient of increasing concentration of imidazole (10, 20, 30, 40mM), the His-tagged proteins were eluted with an imidazole gradient (80500mM imidazole in buffer containing 25mM Tris-HCl, pH-7.5, 300mM NaCl, 10% glycerol, 1mM -ME, 0.25% Tween 20). After elution, the peak protein fractions (in the range of 100250mM imidazole) were dialyzed against Buffer C (1X PBS, pH 7.5, 1mM dithiothreitol (DTT), and 25% glycerol) and stored at 20C in aliquots.

The Corona virus nsp12 (GenBank: MN908947) gene, cloned into a modified pET24b vector, with the C-terminus possessing a 10 His-tag, was a gift from Dr. Whitney Yin. The plasmid was transformed into E. coli BL21 (DE3) RIPL Codon-plus cells, and the transformed cells were cultured at 37C in LB media containing 100mg/L ampicillin. After the OD600 reached 0.8, the culture was cooled to 16C and supplemented with 0.5mM IPTG. After overnight induction, the cells were harvested through centrifugation, and the pellets were re-suspended in lysis buffer (20mM Tris-HCl, pH 8.0, 150mM NaCl, 4mM MgCl2, 10% glycerol). The rest of the procedure is same as above with following modifications: the His-tagged protein was eluted with an imidazole gradient (80250mM imidazole in buffer containing 20mM Tris-HCl, pH 8.0, 150mM NaCl, 4mM MgCl2, 10% glycerol). Similarly, nsp7 and nsp8 genes, individually cloned in pET22b and pET30a+ vectors, respectively, were expressed in E. coli as described in case of NEIL proteins. After elution, the peak protein fractions of these proteins were dialyzed against Buffer D (20mM Tris-HCl, pH 8.0, 250mM NaCl, 1mM DTT, 25% glycerol) and stored at 20C in aliquots.

For assembling the stable nsp12-nsp7-nsp8 complex, purified nsp12 was incubated with nsp7 and nsp8 at 4C for three hours, at a molar ratio of 1: 2: 2 in a buffer containing 20mM Tris-HCl, pH 7.5, 250mM NaCl and 4mM MgCl288.

RdRp assay for CoV-2-5-UTR ZnF-site was conducted using a self-priming RNA oligo and one short RNA oligo was used as the primer for such assay for CoV-2-3-UTR ZnF-site containing sequence as template (Supplementary Table3). Oligos were mixed at the following final concentrations in 20L reaction volume: Tris-HCl (pH 8, 25mM), RNA short primer (200M), RNA template (2M), [32P]-UTP (0.1M), BSA (1mg/ml), 0.1M GTP, CTP, ATP and 0.01M UTP and SARS-CoV-2 RdRp complex (~0.1M) on ice. For NEIL2 binding, the indicated concentrations of NEIL2 were incubated in the buffer with RNA on ice for 15min. Reactions were stopped after 15, 30 or 60min by the addition of 20L of a formamide/EDTA (50mM) mixture and incubated at 95C for 10min. Samples were run in a 8% urea PAGE using 1x Tris-borate-EDTA as the running buffer. The gels were exposed to a Phosphor screen for 46h and images analyzed using a Typhoon FLA 7000 phosphorimager (GE Healthcare).

RNA-EMSAs with full length CoV-2 5- and 3-UTRs were carried out as previously described89 with some modifications. Briefly, the 297-nt long 5- and the 200-nt long 3-UTR RNAs (sequences in Supplementary Table2, synthesized and cloned in plasmids by GenScript Inc.) were synthesized by in vitro transcription and end-labeled with [-32P] ATP. The indicated concentrations of components were mixed in 15l reactions containing 0.3% poly (vinyl alcohol) (Sigma, catalog no. P-8136), 2mM MgCl2, 0.1 U RNase inhibitor (Biobharati Life Science, India), 1mM DTT, 20mM HEPES-NaOH pH 7.5, 150mM NaCl, and 20% glycerol, and incubated at room temperature for 5min. The RNA-protein complexes were resolved on a native gel (4% 89:1 polyacrylamide gel containing 2.5% glycerol, 50mM Tris, and 50mM glycine) at 4C for 90min. Our EMSAs were designed to examine both the affinity (when RNA is in trace amount) and the stoichiometry (when RNA is not in trace amount) of the protein component required to form complexes following principles described before90. Hill coefficient was calculated as described before89. RNA-EMSA with short (38-mer) oligonucleotide (oligo) probes were performed as described before36,91, with some modifications. Sequences of the oligonucleotides are listed in Supplementary Table3. Briefly, [-32P]ATP labeled RNA oligos were incubated with 101000nM of purified protein in a binding buffer containing 10mM Tris-HCl buffer (pH 7.6), 15mM KCl, 5mM MgCl2, 0.1mM DTT, 10 U of RNase inhibitor, 1g BSA, and 0.2mg/ml yeast tRNA in a 1020l reaction volume. After a 10-min incubation at room temperature RNA-protein complexes were resolved on a 5% non-denaturing polyacrylamide gel at 120V using 0.5x Tris-borate-EDTA as the running buffer at 4C. For titration assays with short oligos the reaction mix was prepared without yeast tRNA. Gels were fixed in an Acetone: Methanol: H2O (10:50:40) solution for 10min, exposed to a Phosphor screen for 1216h and scanned using Typhoon FLA 7000 phosphorimager.

Lung tissues from freshly euthanized uninfected and SARS-CoV-2 infected hamsters were used for DNA damage analysis. Genomic DNA was extracted using the Genomic tip 20/G kit (Qiagen) per the manufacturers protocol, to ensure minimal DNA oxidation during the isolation steps. The DNA was quantitated by Pico Green (Molecular Probes) in a black-bottomed 96-well plate and gene-specific LA-qPCR assays were performed as described earlier33,39 using LongAmp Taq DNA Polymerase (New England Biolabs, Catalog no. M0323). The LA-qPCR reaction was set for all genes from the same stock of diluted genomic DNA sample, to avoid variations in PCR amplification during sample preparation. Preliminary optimization of the assays was performed to ensure the linearity of PCR amplification with respect to the number of cycles and DNA concentration (1015ng). The final PCR reaction conditions were optimized at 94C for 30s; (94C for 30s, 5560C for 30s depending on the oligo annealing temperature, 65C for 10min) for 25 cycles; 65C for 10min. Since amplification of a small region is independent of DNA damage, a small DNA fragment (~200500bp) from the corresponding gene(s) was also amplified for normalization of amplification of the large fragment. Primer sequences used in the assay are listed in Supplementary Table1. The amplified products were then visualized on gels and quantitated with ImageJ software (NIH). The extent of damage was calculated in terms of relative band intensity with the uninfected control mice/hamster sample considered as 100.

All statistical tests were performed using R version 3.2.3 (2015-12-10). Standard t-tests were performed using python scipy.stats.ttest_ind package (version 0.19.0) with Welchs Two Sample t-test (unpaired, unequal variance (equal_var=False), and unequal sample size) parameters. Multiple hypothesis correction was performed by adjusting p values with statsmodels.stats.multitest.multipletests (fdr_bh: Benjamini/Hochberg principles). The results were independently validated with R statistical software (R version 3.6.1; 2019-07-05). Pathway analysis of gene lists were carried out via the Reactome database and algorithm. Reactome identifies signaling and metabolic molecules and organizes their relations into biological pathways and processes. Kaplan-Meier analysis was performed using lifelines python package version 0.14.6. Violin and Swarm plots were created using python seaborn package version 0.10.1.

Graph generation and analysis of statistical significance between two sets of data were performed with Microsoft excel, GraphPad Software (https://www.graphpad.com/quickcalcs/pvalue1.cfm) and MedCalc statistical software (https://www.medcalc.org/calc/comparison_of_means.php). p=values<0.05 were considered statistically significant.

Further information on research design is available in theNature Portfolio Reporting Summary linked to this article.

The rest is here:
The DNA glycosylase NEIL2 is protective during SARS-CoV-2 infection - Nature.com