key: cord-0305614-x29qu1g5
authors: Sutter, Sereina O.; Lkharrazi, Anouk; Schraner, Elisabeth M.; Michaelsen, Kevin; Meier, Anita Felicitas; Vogt, Bernd; Büning, Hildegard; Fraefel, Cornel
title: Adeno-associated virus type 2 (AAV2) uncoating is a stepwise process and is linked to structural reorganization of the nucleolus
date: 2021-12-13
journal: bioRxiv
DOI: 10.1101/2021.12.13.472349
sha: c547160f7d7530371b644a55d0982e4125421945
doc_id: 305614
cord_uid: x29qu1g5

Nucleoli are membrane-less structures located within the nucleus and are known to be involved in many cellular functions, including stress response and cell cycle regulation. Besides, many viruses can employ the nucleolus or nucleolar proteins to promote different steps of their life cycle such as replication, transcription and assembly. While adeno-associated virus type 2 (AAV2) capsids have previously been reported to enter the host cell nucleus and accumulate in the nucleolus, both the role of the nucleolus in AAV2 infection, and the viral uncoating mechanism remain elusive. In all prior studies on AAV uncoating, viral capsids and viral genomes were not directly correlated on the single cell level, at least not in absence of a helper virus. To elucidate the properties of the nucleolus during AAV2 infection and to assess viral uncoating on a single cell level, we combined immunofluorescence analysis for detection of intact AAV2 capsids and capsid proteins with fluorescence in situ hybridization for detection of AAV2 genomes. The results of our experiments provide evidence that uncoating of AAV2 particles occurs in a stepwise process that is completed in the nucleolus and supported by alteration of the nucleolar structure. Author Summary Adeno-associated virus (AAV) capsids have been reported to enter the host cell nucleus and accumulate in the nucleolus. However, both the role of the nucleolus in AAV2 infection as well as the viral uncoating mechanism remain unknown. Here, we provide evidence that uncoating of the AAV2 particle is a stepwise process that is completed in the nucleolus and supported by alteration of the nucleolar morphology.

7 146 Interestingly, we did not only observe capsid-positive, genome-negative (AAV2 147 capsid+DNA-) signals in the cytoplasm, as it would be expected when capsids are 148 intact and therefore do not allow binding of the FISH probe to the virus genome, but 149 frequently also capsid-positive, genome-positive signals (AAV2 capsid+DNA+; see 150 insets Fig 1A, 0 h, 3 h, 10 h) . This indicates that either the virus stocks contained 151 improperly encapsidated AAV2 DNA or that the AAV2 genome is accessible to the 152 FISH probe within the cytoplasm (herein referred to as genome accessibility). While 153 the main focus of this study was on simultaneously tracking AAV2 capsids and AAV2

154 genomes in the nucleus, it was important to first investigate the origin of the AAV2 155 capsid-positive and AAV2 genome-positive signals in the cytoplasm. 8 172 Co-detection of AAV2 capsids and AAV2 genomes in the 173 cytoplasm is supported by AAV2 genome accessibility and 174 requires acidification 175 To address the question whether wtAAV2 stocks contained improperly encapsidated 176 AAV2 genomes, wtAAV2 particles were directly applied to fibronectin coated 177 coverslips and processed for IF-FISH (Fig 2A) . While in the untreated wtAAV2

178 samples all capsid-positive signals (green) were genome-negative, only genome-179 positive signals (red) but no capsids were observed upon incubation for 5 min at 180 75°C, which is known to destabilize AAV2 capsids [34] . In the heat-treated samples, 181 the AAV2 capsids were indeed disintegrated as confirmed by electron microscopy 182 (Fig 2B) , and the AAV2 genome signals disappeared upon DNase I treatment ( Fig   183 2A) . These experiments demonstrate that the virus stocks were not contaminated 184 with improperly encapsidated AAV2 DNA. To address the hypothesis that co-185 detection of AAV2 capsids and AAV2 genomes in the cytoplasm is enabled by 186 genome accessibility, we determined the ratios of AAV2 capsid+DNA+/AAV2 187 capsid+DNA-signals at different timepoints after infection using CellProfiler. As 259 Intriguingly, we noticed a distinct difference in the nucleolar structure when 260 comparing AAV2 DNA-positive nucleoli that were positive also for intact AAV2 261 capsids with AAV2 DNA-positive nucleoli that were negative for intact AAV2 capsids 262 or positive for AAV2 capsid proteins. Specifically, the nucleoli appeared dense when 263 positive for both AAV2 DNA and AAV2 capsids and dispersed when positive for 264 AAV2 DNA and negative for AAV2 capsids or positive for AAV2 DNA and AAV2 265 capsid proteins ( Fig 4A, Fig 5B) . Image-based quantification of the mean integrated 266 intensity of AAV2 capsid signals relative to the nucleolar structure revealed a higher 267 capsid signal intensity in dense nucleoli than in dispersed nucleoli ( Fig 4C) . Overall, 268 these experiments imply that complete AAV2 uncoating takes place in the nucleoli 269 and coincides with changes in the nucleolar structure. 356 confirming the efficient G1 arrest (Fig 7A) . Image-based cell cycle analysis showed 357 that the rate of complete uncoating (AAV2 capsid-DNA+/AAV2 capsid+DNA+) was 358 approximately 4-fold lower in the G1-arrested cells compared to the released cells 359 (Fig 7B) . The double thymidine block did not per se influence the rate of complete 360 uncoating, as the ratios of AAV2 capsid-DNA+/AAV2 capsid+DNA+ signals in G1 361 cells were comparable in presence or absence of thymidine ( Fig 7C) . Moreover, 362 neither the blocking with nor the release from thymidine influenced the total area of 363 the nucleoli during cell cycle progression (Fig 7D) . 386 genome signals in the nucleoplasm (Fig 8, A and B ). This shows that complete 387 uncoating can be induced by changes in the nucleolar structure (disruption) even 388 when cells are in G1 phase where normally no efficient complete uncoating is 389 observed (Fig 7) . 479 The exact mechanism that drives step 1 of the AAV2 uncoating process remains to 480 be investigated. However, our data show that it is enhanced by acidification, as co- 

as demonstrated previously [25], and also the AAV2 200 capsid+DNA+/AAV2 capsid+DNA-signal ratios in the cytoplasm (Fig 3, A and B; see 201 also insets in Fig 3A). Collectively, these experiments confirm the specificity of the IF-202 FISH assay and support the hypothesis that the co-localization of AAV2 capsids and directly applied to fibronectin coated coverslips and processed for IF analysis 208 combined with FISH and CLSM. Intact capsids were stained using an antibody that 209 detects a conformational capsid epitope (green). AAV2 DNA (red) was detected with 210 an Alexa Fluor

AAV2 DNA (magenta) was 220 detected with an Alexa Fluor (AF) 647 labeled, amine-modified DNA probe that 221 binds to the AAV2 genome. (A) Genome accessibility of AAV2 capsids after 222 inhibition of the endosome-lysosome system acidification

Image-based quantification of the 224 genome accessibility (ratio of AAV2 capsid+DNA+/AAV2 capsid+DNA-signal) of 50 , we continued to analyze the distribution of AAV2 capsids and genomes in 234 the nuclei of individual cells. For this, NHF cells were mock-infected or infected with 235 wtAAV2 (MOI 20`000) and 24 h later processed for combined IF-FISH and CLSM to 236 detect AAV2 capsids and genomes. Interestingly, we observed three distinct 237 patterns of nucleolar AAV2 genome and AAV2 capsid staining: (I) nucleoli with 238 robust AAV2 genome and AAV2 capsid signal, (II) nucleoli with robust AAV2 DNA 239 signal but weak AAV2 capsid signal, and (III) nucleoli in which only the viral DNA 240 was detected in absence of capsids (Fig 4, A and B, see also S1 movie). The pattern

If the absence of AAV2 capsid staining in the nucleoli with positive AAV2 genome 245 signal was indeed due to complete viral uncoating, we would expect the presence of 246 disassembled AAV2 capsid proteins in those nucleoli. To assess this hypothesis, 247 NHF cells were mock-infected or infected with wtAAV2 (MOI 20`000) and 24 h later 400 nucleolar capsid

To further assess whether capsid disassembly overlaps with cell cycle progression, 406 NHF cells were either mock-infected or infected with wtAAV2 (MOI 20`000). 24 h 407 later, the cells were fixed and processed for IF-FISH, CLSM and image-based cell 408 cycle analysis and quantification. Specifically, we used the DAPI integrated intensity 409 protocol to determine the cell cycle phase and the IF-FISH protocol to detect intact 410 AAV2 capsids, the disassembled AAV2 VP1/2/3 capsid proteins, and the AAV2 DNA

capsids (green) or capsid proteins (yellow) were detected using either an antibody 424 against intact AAV2 capsids (conformational capsid epitope) or an antibody (linear 425 epitope) against VP1, VP2 and VP3. AAV2 DNA (magenta) was detected with an 18

AF) 647 labeled, amine-modified DNA probe that binds to the AAV2 427 genome. Nuclei were counterstained with DAPI and illustrated as white lines

Nucleoli are membrane-less and dynamic subnuclear structures, which were mainly 433 known for their role in ribosome biosynthesis. However, nucleoli have a function in 434 numerous other cellular processes as well, such as cell cycle regulation, stress 435 response and apoptosis

Many different viruses can exploit the nucleolus or nucleolar proteins to drive different 439 steps of their life cycle including replication, transcription, and assembly

Similarly, nucleolar upstream binding factor (UBF) and nucleophosmin (B23.1) are 444 recruited to adenovirus replication compartments to promote viral DNA replication

The autonomous parvovirus minute virus of mice has been shown to replicate its 446 DNA in the nucleoli of mouse fibroblasts

Additionally, borna disease virus 447 transcription and replication take place in the nucleoli as well

Moreover, specific 448 mRNAs and proteins of many different viruses, including HIV-1, Japanese encephalitis 449 virus, and Semliki Forest virus traffic through the nucleolus for processing, and the 450 inhibition of such trafficking affects virus replication

Cells and viruses 543 Normal human fibroblast (NHF) cells were kindly provided by X.O. Breakefield 544 (Massachusetts General Hospital

USA) were maintained in 548 growth medium containing Dulbecco's modified Eagle medium (DMEM) 549 supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin G, 100 µg/ml 550 streptomycin, and 0.25 µg/ml amphotericin B (1% AB) at 37°C in a 95% air-5% CO 2 551 atmosphere

Germany) and the Viral Vector Facility (VVF) of the 23

The VP1 AAV2 mutant ( 76 HD/AN) was 554 constructed according to Girod

76 HD/AN mutant construct was generated by mutating two key residues 556 76 HD to 76 AN using K-76 HD/AN (5

The following primary antibodies were used: anti-AAV2 intact particle

anti-AAV VP1/VP2/VP3 (VP51, ProGen, 563 dilution for IF 1:10), anti-AAV VP1/VP2 (A69, ProGen, dilution for IF 1:10)

Abcam 565 ab5821; dilution for IF 1:200), anti-cyclin A (Santa Cruz sc-751

The following secondary antibodies were used: Alexa Fluor 594 goat anti-rabbit IgG 567 (Life Technologies A11037, dilution for IF 1:500), Alexa Fluor 488 goat anti-mouse IgG

Viral infection and treatments

HDFn or Vero cells were seeded onto coverslips (12-mm diameter

The next day the cells were washed 573 with PBS and either mock-infected or infected with wtAAV2 at a multiplicity of 574 infection (MOI) of 20`000 genome containing particles (gcp) per cell in 250 µl of 575 DMEM (0% FBS, 1% AB) pre-cooled to 4°C. The plates were first incubated for 576 30 min at 4°C to synchronize viral uptake and then incubated at 37°C in a humidified 577 95% air-5% CO 2 incubator for the indicated time period. For acidification experiments 24 this, cells were 581 seeded in 10 cm tissue culture dishes (5x10 5 cells per dish) and 12 h later the growth 582 medium was replaced with medium (DMEM, 10% FBS, 1% AB) containing 3 mM 583 thymidine. After 12 h of incubation, the cells were washed with PBS, trypsinized and 584 split at a density of 6x10 4 cells per well into 6-well tissue culture plates containing 585 coverslips. In order to complete the double thymidine block, the growth medium was 586 replaced 12 hours later by medium containing 3 mM thymidine

Briefly, NHF cells were seeded onto coverslips (12-mm 595 diameter

The next day, the cells were washed 597 with PBS, processed as indicated in the results and the figure legends, counterstained 598 with DAPI and imaged by confocal laser scanning microscopy

An automated CellProfiler (V.2.2.0-V.4.0.7) 600 pipeline measured the integrated intensity of DAPI. Next, the histograms of DAPI, 601 corresponding to the DNA content, were plotted and visual thresholds for each cell 602 cycle phase were selected

To validate the DAPI integrated intensity protocol, NHF cells were synchronized using 608 a double thymidine block (as described above)

5 ml ice-cold 100% ethanol, 612 centrifuged, washed once again with PBS and stained with a freshly made solution 613 containing 1 µg/mL DAPI, 0.05% Triton X-100 and 0.1 mg/mL ribonuclease A

After incubation, 615 the cells were washed twice with PBS and then resuspended in 200 µl PBS prior to 616 analysis (SONY SP6800 Spectral Analyzer). For confocal laser scanning microscopy, 617 the coverslips were embedded in ProLong Anti-Fade mountant with DAPI (Molecular 618 Probes

Images were analyzed as described in 620 the section cell cycle analysis based on 4`,6-diamidino-2-phenylindole (DAPI) staining

FISH was performed essentially as described previously by

9-kb DNA fragment containing the AAV2 genome without the inverted terminal 625 repeats was amplified by PCR from plasmid pDG using forward (5`-626 CGGGGTTTTACGAGATTGTG-3`) and reverse

and the following conditions: 30 s at 95°C; 35 cycles of 10 s at 98°C

at 58°C, and 75 s at 72°C; and 10 min at 72°C. The PCR sample was then digested 629 with DpnI to cut the residual template DNA and purified with the Pure Link PCR 630 purification kit

24 hours after infection, the cells were washed with PBS, fixed for 30 637 min at room temperature (RT) with 2% PFA

The cells were then quenched for 10 min with 50 mM NH 4 Cl

blocked for 10 min 640 with 0.2% gelatin (in PBS) followed by two washing steps with PBS before blocking for 641 30 min in PBST (0.05% Tween 20 in PBS) supplemented with 3% BSA at 4°C. After 642 antibody staining in PBST-BSA (3%, 25 µl/coverslip) for 1h at RT in the dark in a 643 humidified chamber, the cells were washed three times for 5 min with PBST (0.1%), 644 post-fixed with 2% PFA and blocked with 50 mM glycine in PBS

Hybridization solution (20 l per coverslip) containing 1 ng/l of the labeled DNA probe, 646 50% formamide, 7.3% (w/v) dextran sulfate, 15 ng/l salmon sperm DNA

The coverslips with the fixed and permeabilized cells 649 facing down were placed onto a drop (20 l) of the denatured hybridization solution 650 and incubated overnight at 37°C in a humidified chamber (note that the cells were not 651 denatured, as the AAV2 genome is present as ssDNA). The next day, the coverslips 652 were washed three times with 2x SSC at 37°C, three times with 0.1x SSC at 60°C, and 653 twice with PBS at RT. To confirm the FISH signal

results) were treated with DNase I (1U/ µl) for 1 h at 37°C followed by inactivation in 655 30% formamide, 0,1% Triton-X 100 and 2x SSC for 10 min at RT

The cells were then embedded in ProLong Anti-Fade mountant with or without DAPI 657 (Molecular Probes

To prevent cross talk Bitplane, Oxford 662 Instruments

For this, 10 µl of the wtAAV2 stock were placed onto a parafilm strip and 666 adsorbed to carbon coated parlodion films mounted on 300 mesh/inch copper grids by 667 placing upside-down on the drop and incubated for 10min at RT. Washing was done 668 by transferring the grid to a H 2 O drop. Subsequently the grid was placed onto a drop 669 of phosphotungstic acid (PTA, pH 7.0) for 60 seconds. Remaining liquid was removed 670 by tipping the edge of the grid on a filter paper. The samples were analyzed in a Philips 671 CM 12 transmission electron microscope

For image-based quantification and data analysis, at least 50 individual cells per 676 sample or condition were recorded and analyzed using different CellProfiler

7) pipelines. The output csv-files were further analyzed using Matlab

deviation (SD), statistical analysis of individual cells was either performed by unpaired

Student`s t-test or an unpaired t-test with Welch`s correction (not assuming equal

If not stated otherwise, each graph illustrates one representative experiment

Targeted 685 integration of adeno-associated virus (AAV) into human chromosome 19

Complementary Adeno-Associated Viral Serotype Vectors in Murine and Nonhuman 690 Primate Models

Herpes simplex virus types 1 and 2 692 completely help adenovirus-associated virus replication

Nucleotide sequence and organization of the 695 adeno-associated virus 2 genome

Spliced adenovirus-associated virus RNA

The adeno-associated virus (AAV) Rep 700 protein acts as both a repressor and an activator to regulate AAV transcription during 29 701 a productive infection

The 704 assembly-activating protein promotes capsid assembly of different adeno-associated 705 virus serotypes

Comprehensive AAV capsid fitness 708 landscape reveals a viral gene and enables machine-guided design

Intracellular transport of recombinant adeno-713 associated virus vectors

An 715 essential receptor for adeno-associated virus infection

Adeno-associated virus 2 infection requires 718 endocytosis through the CLIC/GEEC pathway

Intrinsic 721 phospholipase A2 activity of adeno-associated virus is involved in endosomal escape 722 of incoming particles

A 110-kDa nuclear shuttle protein, nucleolin, specifically binds 724 to adeno-associated virus type 2 (AAV-2) capsid

Adeno-associated virus interactions with B23/Nucleophosmin: identification of sub-728 nucleolar virion regions

The nucleolus: an old factory with unexpected 731 capabilities

Emerging concepts of 733 nucleolar assembly

Nucleolar Stress: hallmarks, sensing mechanism and 735 diseases

The nucleolus under 737 stress

Viruses and the nucleolus: the fatal attraction. Biochimica et 741 biophysica acta

Enhancement of adeno-associated virus infection by 743 mobilizing capsids into and out of the nucleolus

Detection and Localization of a Class of Proteins Immunologically Related to a 747 100-kDa Nucleolar Protein

Infectious entry pathway of adeno-753 associated virus and adeno-associated virus vectors

Super-resolution 756 imaging of nuclear import of adeno-associated virus in live cells

Green 759 fluorescent protein-tagged adeno-associated virus particles allow the study of 760 cytosolic and nuclear trafficking

Infectious entry pathway of adeno-763 associated virus and adeno-associated virus vectors

Adenovirus

Facilitated Nuclear Translocation of Adeno-Associated Virus Type 2

Differential 769 internalization and nuclear uncoating of self-complementary adeno-associated virus 770 pseudotype vectors as determinants of cardiac cell transduction

Nuclear 773 Transport and Uncoating Limit Recombinant Adeno-Associated Virus

Subcellular 777 compartmentalization of adeno-associated virus type 2 assembly

The VP1 capsid 780 protein of adeno-associated virus type 2 is carrying a phospholipase A2 domain 781 required for virus infectivity

Comparative Analysis of Adeno-Associated Virus Capsid Stability and Dynamics

Latonen L. Phase-to-Phase With Nucleoli -Stress Responses, Protein 790 Aggregation and Novel Roles of RNA

NOPdb: Nucleolar 793 Proteome Database-2008 update

Nucleolus as 796 an emerging hub in maintenance of genome stability and cancer pathogenesis

The nucleolus -a gateway to viral infection?

RNA viruses: hijacking the dynamic nucleolus

Nucleolin is 803 required for an efficient herpes simplex virus type 1 infection

Relationship between adenovirus DNA 806 replication proteins and nucleolar proteins B23.1 and B23.2

Nucleolar protein upstream binding factor 809 is sequestered into adenovirus DNA replication centres during infection without 34

RNA polymerase I location or ablating rRNA synthesis

Identification of 813 nucleophosmin/B23, an acidic nucleolar protein, as a stimulatory factor for in vitro 814 replication of adenovirus DNA

Interactions of minute virus of mice and 817 adenovirus with host nucleoli

Adenovirus and minute virus of mice DNAs are 820 localized at the nuclear periphery

The nucleolus is the site of Borna disease 823 virus RNA transcription and replication

Ribozyme-mediated inhibition of HIV 825 1 suggests nucleolar trafficking of HIV-1 RNA

Nuclear 828 Localization of Japanese Encephalitis Virus Core Protein Enhances Viral

A Single Amino Acid Change 831 in the Nuclear Localization Sequence of the nsP2 Protein Affects the Neurovirulence 832 of Semliki Forest Virus

Separate Basic Region Motifs within the

Adeno-Associated Virus Capsid Proteins Are Essential for Infectivity and Assembly

Adeno-associated to passage through the cytoplasm and are maintained until uncoating occurs in 839 the nucleus

Characterization 841 of AAV vector particle stability at the single-capsid level

Adeno-Associated Virus Activates an Innate Immune 844 Response in Normal Human Cells but Not in Osteosarcoma Cells

Breaking Bad: How Viruses Subvert the Cell 847 Cycle

Uncoating of HIV-1 requires cellular activation

Nucleolin, the major nucleolar protein of 855 growing eukaryotic cells: an unusual protein structure revealed by the nucleotide 856 sequence

HSV-1) Gives Rise to a Mosaic of Cells 861 Replicating either AAV2 or HSV-1

Capsid 864 Engineering Overcomes Barriers Toward Adeno-Associated Virus Vector-Mediated 865 Transduction of Endothelial Cells

A recombinant plasmid from which an 868 infectious adeno-associated virus genome can be excised in vitro and its use to study 869 viral replication

Cell cycle staging of individual cells 871 by fluorescence microscopy

At 5 hpi, the samples were processed for IF-FISH 878 and CLSM. Intact capsids were stained using an antibody that detects a 879 conformational capsid epitope (green). AAV2 DNA (red) was detected with an Alexa

Fluor (AF) 647 labeled, amine-modified DNA probe that binds to the AAV2 genome

Nuclei were counterstained with DAPI (blue)

At 24 885 hpi, the cells were fixed and processed for multicolor IF analysis combined with 886 FISH and CLSM. Intact capsids (green) or capsid proteins (yellow) were detected 887 using either an antibody against intact AAV2 capsids (conformational capsid 888 epitope) or an antibody (linear epitope) against VP1, VP2 and VP3. AAV2 DNA (red) 889 was detected with an Alexa Fluor (AF) 647 labeled, amine-modified DNA probe

CLSM and quantification of the nucleolar structure 895 of 100 individual nuclei in mock-or AAV2 infected cells. p-values were calculated 896 using an unpaired Student`s t-test (* -p ≤ 0.05, ** -. (2) Nuclei and cyclin A stainings were identified as primary objects in

Histograms of the DAPI integrated 904 intensities were plotted using an automated script in Matlab and visual thresholds 905 were set. (6) Cells were classified in CellProfiler using the visual thresholds obtained 906 in step 5. (7) The classification of each nucleus into G1, S or G2 was overlayed on thymidine (3 mM) block. After the release, the cells were either mock-treated at least 5`000 scored events. CLSM analysis shows the mean value of three 915 experiments

Capsid disassembly coincides with cell cycle progression in HDFn 926 cells. HDFn cells were infected with wtAAV2 (MOI 20`000)

capsids (green) or capsid proteins (yellow) were detected using either an antibody 929 against intact AAV2 capsids (conformational capsid epitope) or an antibody (linear 930 epitope) against VP1, VP2 and VP3. AAV2 DNA (magenta) was detected with an

AF) 647 labeled, amine-modified DNA probe that binds to the AAV2 932 genome. Nuclei were counterstained with DAPI and illustrated as white lines

Quantification of at least 50 nuclei positive for intact AAV2 capsids or capsid proteins mock-infected or infected with wtAAV2 (MOI 10`000), HSV-1 (MOI 0.5) or co-939 infected with wtAAV2 (MOI 5`000) and HSV-1 (MOI 0.5)

Nucleoli (Nuc) were visualized using an antibody against fibrillarin (yellow)

At 24 hpi, the cells were fixed 949 and processed for multicolor IF analysis combined with FISH and CLSM. Nucleoli 950 were visualized using an antibody against fibrillarin (yellow)

AAV2 DNA (red) was detected with an Alexa Fluor (AF) 647 labeled