We describe a protocol utilizing fluorescence in situ hybridization (FISH) to visualize multiple herpesviral RNAs within lytically infected human cells, either in suspension or adherent. This protocol includes quantification of fluorescence producing a nucleocytoplasmic ratio and can be extended for simultaneous visualization of host and viral proteins with immunofluorescence (IF).
Mechanistic insight arrives from careful study and quantification of specific RNAs and proteins. The relative locations of these biomolecules throughout the cell at specific times can be captured with fluorescence in situ hybridization (FISH) and immunofluorescence (IF). During lytic herpesvirus infection, the virus hijacks the host cell to preferentially express viral genes, causing changes in cell morphology and behavior of biomolecules. Lytic activities are centered in nuclear factories, termed viral replication compartments, which are discernable only with FISH and IF. Here we describe an adaptable protocol of RNA FISH and IF techniques for Kaposi’s sarcoma-associated herpesvirus (KSHV)-infected cells, both adherent and in suspension. The method includes steps for the development of specific anti-sense oligonucleotides, double RNA FISH, RNA FISH with IF, and quantitative calculations of fluorescence intensities. This protocol has been successfully applied to multiple cell types, uninfected cells, latent cells, lytic cells, time-courses, and cells treated with inhibitors to analyze the spatiotemporal activities of specific RNAs and proteins from both the human host and KSHV.
In their lytic (active) phase, herpesviruses hijack the host cell, causing changes in cell morphology and localization of biological molecules, to produce virions. The base of operations is the nucleus, where the double-stranded DNA viral genome is replicated and packaged into a protein shell, called a capsid1. To begin, the virus expresses its own proteins, hijacking host machinery and preventing expression of non-essential host genes, a process termed the host shutoff effect. The majority of this activity is localized to specific 4′,6-diamidino-2-phenylindole (DAPI)-free nuclear regions called viral replication compartments, comprised of both host and viral proteins, RNAs, and viral DNA2. The cell is overhauled to provide space and resources for the replication compartments and thus assembly of viral capsids. Once the capsid exits the nucleus, how the capsid is enveloped in the cytoplasm to produce a membrane-bound viral particle, also known as a virion, is unclear. Understanding of the localization and spatial shifts of both host and viral biomolecules during the lytic phase provides deeper mechanistic insight into the arrangement of the replication compartment, host shutoff effect, the virion-egress pathway, and other processes related to herpesviral infection and replication.
Currently the best method to detect and study these changes is the visualization of proteins and RNAs in infected cells with immunofluorescence (IF) and fluorescent in situ hybridization (FISH), respectively. Use of a time-course with these techniques reveals the localization of biomolecules at key points of the lytic phase or simply, spatiotemporal data. FISH and IF complement other biochemical techniques, such as inhibition of a cellular process (e.g., inhibition of viral DNA replication), RT-qPCR (real-time polymerase chain reaction), RNA sequencing, Northern blots, mass spectrometry, Western blotting, and analysis of viral DNA production, that may provide a more global picture of cellular activities.
We developed RNA FISH strategies to examine the RNA products from specific genes and a computational analysis that quantitatively calculates the nucleocytoplasmic ratio of a specific gene product. The sample preparation, modified from earlier publications by Steitz and colleagues3,4, is relatively easy and can be used for both adherent and suspended cells. The protocol is also adaptable for simultaneous use of multiple RNA FISH strategies (double RNA FISH) or RNA FISH with IF strategies. Development of a specific FISH strategy is challenging, but suggestions to improve success are outlined. The data analysis described here is quantitative if fluorescent beads and strong markers of compartment boundaries are used and offers additional insight into the micrographs, insight that removes observation bias. The detailed protocol is designed for both latent and lytic cells infected by Kaposi's sarcoma-associated herpesvirus (KSHV) and can be used with uninfected cells or cells infected by other herpesviruses5. The methods of quantitation are applicable to studies on nucleocytoplasmic shifts or relocalization between subcellular compartments in most cells.
1. Design of fluorescence in situ (FISH) anti-sense oligonucleotides to detect a specific herpesviral transcript
2. Oligonucleotide and cell preparation
3. Fixation, Immunofluorescence (Optional), Hybridization, and Visualization of Viral RNAs
4. Quantification of FISH and IF images to highlight subcellular localization and to determine nucleocytoplasmic ratio of fluorescence
The FISH and IF methods detailed in this manuscript are shown in Figure 1 along with the quantification of results by line traces of fluorescent intensity. The results presented here are semi-quantitative and offer insight into localization, rather than into comparisons between intensities of different fluorescent stains because experiments did not include a fluorescent bead in the slide preparation. Figure 1 also reveals that the cytoplasmic and nuclear areas and their ratios are different for latent and lytic KSHV-infected cells. Thus, area is controlled in the nucleocytoplasmic ratio showcased in Figure 2. Figure 2 validates the calculation detailed in this manuscript for a nucleocytoplasmic ratio with the use of a nuclear control, the viral polyadenylated nuclear (PAN) RNA, and a cytoplasmic control, the host GAPDH mRNA. Figure 3 reveals that when KSHV DNA replication is inhibited in the lytic phase by the use of either phosphonoacetic acid (Doxy + PAA) or cidofovir (Doxy + Cido), the early ORF59-58 transcript shifts to a predominantly cytoplasmic localization. The micrographs and the two quantification methods in Figure 3 support this result and reveal that PAN RNA localizes to specific nuclear sites despite inhibition of viral DNA replication and the change seen for early ORF59-58 transcript.
Figure 1: Lines traces of fluorescent intensity reveal subtleties in fluorescence in situ hybridization (FISH) of KSHV transcripts and immunofluorescence (IF) of KSHV replication compartments. (A–B) Confocal images of TREx RTA (tetracycline inducible viral replication and transcription activator protein) BCBL-1 cells12 that have been induced into the lytic phase for 24 h with doxycycline (Doxy). Scale bar indicates 10 µm. (A) Fluorescence in situ hybridization (FISH) for viral RNAs (green) and immunofluorescence (IF) for viral single-stranded DNA binding protein (ORF6/SSB) (red), a component of KSHV replications compartments, reveal that viral transcripts localize in the cytoplasm, nucleus, and in nuclear foci outside ORF6/SSB enriched areas, also known as replication compartments. The anti-SSB antibody10 was diluted to 1:200 in 0.4% BSA/1x PBS and detected with 1:500 anti-rabbit Alexa Fluor 594 secondary antibody in 0.4% BSA/1x PBS. All anti-sense oligonucleotides used throughout this study are provided in Table 1. The detection of ORF59-58 mRNA includes both the bicistronic and monocistronic transcripts. However, in KSHV-infected JSC-1 cells, the monocistronic mRNA is at least 18-fold less abundant than the bicistronic transcript and likely contributes only a minor portion of the total fluorescent signal observed13. Moreover one of the PAN RNA oligonucleotides (SB88) can also detect the viral transcript for K7. The signal from a detection of K7 will not be as significant compared to the signal detecting KSHV PAN RNA, which is present at nearly 80% of all polyadenylated RNA in a lytic KSHV-infected cell14. Additionally one of the four anti-sense oligonucleotides (tkv13) in the detection of the K8.1 mRNA is able to bind to multiple isoforms of K8.1 and other isoforms of nearby open reading frames (ORF). The FISH signal from only oligonucleotide tkv13 is insufficient (data not shown). The combined hybridization of the four oligonucleotides and the binding of them on the same transcript likely provides the observed strong signal. White lines flanking cells in (A) depict the line path of fluorescence intensities for the FISH and IF signals, plotted in (C). (B) Digitally zoomed images of cells in (A) flanked by white lines. For simplicity, the blue DAPI channel is omitted. (C) The plots show the relative fluorescent intensities for each stain along the same line: αSSB (red), viral transcripts (green; transcript indicated on plot), and DAPI (blue). Shaded areas indicate DAPI-reduced regions that correspond to viral replication compartments or SSB/ORF6-enriched areas. (D) The ratio of nuclear area to cellular area changes and thus the fluorescence intensity ratio used throughout was normalized for area. (E) Nuclear and cellular areas measured for TREx RTA BCBL-1 cells with and without undergoing lytic activation. Statistically significant changes are seen compared to uninduced cells. The box and whisker plots represent the 10 and 90 percentiles. Figure reprinted with slight modifications from Vallery, Withers, and colleagues15 under a Creative Commons Attribution license. Please click here to view a larger version of this figure.
Figure 2: Control nuclear and cytoplasmic FISH strategies validate the calculation method of the nucleocytoplasmic ratio. (A) FISH for the host GAPDH mRNA (red) and for the viral polyadenylated nuclear (PAN) lncRNA (green) and DAPI nuclear staining (blue) are positive FISH controls for the calculation method determining the nucleocytoplasmic ratio. Host GAPDH mRNA is a canonical target of the KSHV's host shutoff effect and is degraded upon lytic induction as shown here. (B) Fluorescence intensities along a line indicated by white lines flanking lytic cells in (A). DAPI (blue), PAN RNA (green), and GAPDH mRNA (red). Shaded areas are as defined in Figure 1. (C) Quantification of the fluorescence intensities of cells represented by (A) (n = 150 for each GAPDH sample, n = 75 for the ORF59-58 or K8.1 samples) were performed for three biological replicates of cells shown in Figure 2 and Figure 3. P-values: >0.05 (ns), <0.05 (*), <0.005 (**), and <0.0005 (***). (D) Representative Northern blot of RNA from TREx RTA BCBL-1 cells 24 h after Doxy. The box and whisker plots represent the 10 and 90 percentile. Figure reprinted with slight modifications from Vallery, Withers, and colleagues15 under a Creative Commons Attribution license. Please click here to view a larger version of this figure.
Figure 3: Line traces and calculation of nucleocytoplasmic ratios reveal a strong shift to the cytoplasm for the early lytic ORF59-58 transcript upon inhibition of viral DNA replication. TREx RTA BCBL-1 cells were treated for 24 h with no drug (Unind), doxycycline only (Doxy), or with doxycycline and one inhibitor of herpesviral DNA replication, phosphonoacetic acid (Doxy + PAA) or cidofovir (Doxy + Cido). Panels (A–C) show data from samples collected from three biological replicates. (A) qPCR values for viral intracellular DNA during inhibition of viral DNA replication were normalized to the quantity of promoter DNA of the host-cell GAPDH gene. (B) Northern blot (left) and quantification (right) show total RNA levels during inhibition of viral DNA replication. Uninduced levels of all RNAs were undetectable. (C) Representative FISH images for viral ORF59-58 transcripts (green) and PAN RNA (red) upon inhibition of viral DNA replication. DAPI (blue) was the nuclear stain. (D) Quantification of the fluorescence intensities of cells represented by (C) (n = 75 each) was done on biological triplicates. (E) Fluorescence intensities along lines drawn across cells indicated by white lines in (C) are shown: DAPI (blue), PAN RNA (red), and ORF59-58 mRNA (green). P-values: >0.05 (ns), <0.05 (*), <0.005 (**), and <0.0005 (***). The sequences of all oligonucleotides in this study are provided in Table 1. The box and whisker plots represent the 10 and 90 percentiles. Figure reprinted from Vallery, Withers, and colleagues15 under a Creative Commons Attribution license. Please click here to view a larger version of this figure.
Northern Oligos | |||||||
Oligo No. | Gene | Sequence | Position within the gene | Position with the Reference NC009333.1 Genome (Numbers do not reflect direction or strand) | |||
KORF50 | KSHV RTA/ORF50 | CGCATTGCGGTGGTTGAAATTGCTGG | 1284 to 1309 | 73936 to 73961 | |||
JBW249 | KSHV SOX/ORF37 | TAACCTGACACCACCAAACACACGGTCCAC | 262 to 291 | 57633 to 57662 | |||
tkv379 | KSHV ORF59-58 | TGGAGTCCGGTATAGAATCGGGAACCT | 941 to 967 (ORF59 ORF) | 95879 to 95905 | |||
tkv13 | KSHV K8.1 | AAGGCATAGGATTAGGAGCGCCACAGGGATTTCTGTGCGAAT | 16 to 57 | 76029 to 76070 | |||
SB2 | KSHV PAN RNA | ACAAATGCCACCTCACTTTGTCGC | 664 to 687 | 29496 to 29519 | |||
Rnase P | Human RNase P | TGGGCGGAGGAGAGTAGTCTG | 319 to 339 | N/A | |||
FISH Probes | |||||||
Oligo No. | Gene | Sequence | |||||
SB2 | KSHV PAN RNA | ACAAATGCCACCTCACTTTGTCGC | 664 to 687 | 29496 to 29519 | |||
SB85 | KSHV PAN RNA | CGCTGCTTTCCTTTCACATT | 373 to 392 | 29205 to 29224 | |||
SB88 | KSHV PAN RNA | GTGAAGCGGCAGCCAAGGTGACTGG | 1 to 22 | 28830 to 28854 | |||
tkv13 | KSHV K8.1 | AAGGCATAGGATTAGGAGCGCCACAGGGATTTCTGTGCGAAT | 16 to 57 | 76029 to 76070 | |||
tkv14 | KSHV K8.1 | TGATATTAAGGCATCGGTCAGTTCTGTGGTGGCCTGGA | 377 to 414 | 76390 to 76427 | |||
tkv15 | KSHV K8.1 | GTAAGGTTACGCTTTATCCCTACACACCGACGGTTTACCC | 461 to 500 | 76474 to 76513 | |||
tkv16 | KSHV K8.1 | GGACAAGTCCCAGCAATAAACCCACAGCCCATAGTATG | 688 to 725 | 76701 to 76738 | |||
tkv376 | KSHV ORF59-58 | TAATGTGTTCATTGACCCTCCTGATT | 54 to 79 | 96767 to 96792 | |||
tkv377 | KSHV ORF59-58 | GCCGATCCGTGCACTTGCACTACTCCGGTT | 93 to 122 | 96724 to 96753 | |||
tkv378 | KSHV ORF59-58 | AAGGCTATGCCAGCGTCGAGTACATTCGCA | 300 to 329 | 96517 to 96546 | |||
tkv379 | KSHV ORF59-58 | TGGAGTCCGGTATAGAATCGGGAACCT | 941 to 967 | 95879 to 95905 | |||
tkv380 | KSHV ORF59-58 | AAAGAGTGTGAACGAGTACAGGGCCTT | 1289 to 1315 | 95531 to 95557 | |||
tkv381 | KSHV ORF59-58 | AAACACTGCTGACGCGCAGATCCATTCC | 1423 to 1450 | 95396 to 95423 | |||
tkv382 | KSHV ORF59-58 | TACCTGTGTACTATTGGCGGCGCCTGATACAC | 1571 to 1602 | 95244 to 95275 | |||
tkv383 | KSHV ORF59-58 | GGGTCGAGATTCAGCTAATTAGGCGAAAACTCCACAGG | 2136 to 2173 | 94673 to 94710 | |||
Stellaris | GAPDH | Premade by Stellaris | N/A | N/A | |||
qPCR Primers | |||||||
tkv458 | GAPDH Promoter | CTGCACCACCAACTGCTTAG | N/A | N/A | |||
tkv459 | GAPDH Promoter | GTCTTCTGGGTGGCAGTGAT | N/A | N/A | |||
tkv319 | KSHV ORF39 (gM) | GTGAGGTGCTTCGCTGAGTT | N/A | 60075 to 60094 | |||
tkv320 | KSHV ORF39 (gM) | CCTGGGTCAAGCTGTTGTTT | N/A | 60218 to 60237 | |||
RT-qPCR Primers | |||||||
tkv 455 | K8/K-bZIP Forward RT qPCR Primer | CGAAAGCAAGGCAGATACG | 655 to 673 | 75603 to 75621 | |||
tkv 456 | K8/K-bZIP Reverse RT qPCR Primer for unspliced | GCCATTGTTCCCATTTGAGT | 755 to 774 | 75703 to 75722 | |||
tkv 457 | K8/K-bZIP Reverse RT qPCR Primer for spliced | CATCAGCATGTCGCGAAG | 871 to 888 | 75819 to 75836 | |||
JBW479 | Human RNase P Forward | AGCTTGGAACAGACTCACGG | 238 to 257 | N/A | |||
JBW480 | Human RNase P Reverse | GCGGAGGAGAGTAGTCTGAA | 317 to 336 | N/A |
Table 1: All oligonucleotides used in the analyses of this publication. Table 1 was reproduced with permission from the American Society for Microbiology under a Creative Commons Attribution license from Vallery et al.15.
The protocol described in this report can be adapted to different cell types and includes steps for double RNA FISH and RNA FISH with IF using both monoclonal and polyclonal primary antibodies. Although prepared slides are typically imaged with a confocal microscope, imaging can be performed with a STED (stimulated emission depletion) microscope after modifications of increased antibody concentration and a different mounting medium. For enhanced analysis of individual cells, samples prepared with this protocol may also be sorted, imaged, and analyzed by a cell sorter or flow cytometer with modest changes, as shown by Borah and colleagues16. This protocol, however, cannot be adopted for live cell imaging.
The quantification methods detailed eliminate observation bias and serve to validate a potential nucleocytoplasmic shift. The nucleocytoplasmic ratio also pinpoints when a biomolecule adjusts from being evenly dispersed to localizing in a specific subcellular compartment. The results presented here are semi-quantitative while the protocol outlines ways to strengthen the quantification. The strength of the nucleocytoplasmic ratio and line traces (step 4) depend on the use of fluorescent beads as intensity controls (step 3.13) and the use of clear subcellular markers, such as one for nuclear Lamin A/C. At this time, a clear boundary marker does not exist for KSHV viral replication compartments. Regardless, this calculation can be extended to other subcellular compartments with the use of appropriate markers.
The major hurdle for the protocol detailed in this report is the development of FISH strategies for specific transcripts (step 1). Success relies on the abundance and binding strength of anti-sense oligonucleotides. Specificity to particular transcripts is made even more difficult by the presence of overlapping open reading frames (ORFs) in viral genomes. Thus, viral transcripts often have sequence similarity17 with other viral transcripts from the same genomic region, especially in case of herpesviruses. Often development of a FISH strategy must take advantage of more abundant transcripts. To troubleshoot a lack of a FISH signal, users should perform the FISH protocol with the U2 snRNA FISH strategy to confirm that techniques in human cells and preparation of reagents are adequate. Likewise, the KSHV PAN RNA FISH strategy can confirm lytic activation in KSHV-infected cells. To troubleshoot binding by the anti-sense oligonucleotides, the authors recommend developing several anti-sense oligonucleotides. If all fails, a commercial option is available as demonstrated by the use of the GAPDH FISH strategy in Figure 2 and by Vallery, Withers, and colleagues15.
Stronger algorithms to define the cellular and subcellular boundaries would further eliminate quantification bias. Some analytical image processing software can set boundaries for the cell, nucleus, and more, but require definitive markers. Unusual cell morphologies such as viral replication compartments are difficult for such software – a challenge for future development. Moreover the quantification methods described here are limited to one optical slice of a cell (2D image analysis). While 3D image acquisition is possible18, future development of a quantitative 3D image analysis may provide further insight into spatiotemporal regulation of viral replication compartments.
The authors have nothing to disclose.
We thank Jonathan Rodenfels, Kazimierz Tycowski, and Johanna B. Withers for advice on data analysis. We also thank G. Hayward for the anti-SSB antibody. This work was supported by grants T32GM007223 and T32AI055403 from the National Institutes of Health (to TKV) and NIH grant (CA16038) (to JAS). JAS is an investigator of the Howard Hughes Medical Institute. Figures 1-3 and Table 1 were reproduced with permission from the American Society for Microbiology under a Creative Commons Attribution license from the following publication: Vallery, T. K., Withers, J. B., Andoh, J. A., Steitz, J. A. Kaposi's Sarcoma-Associated Herpesvirus mRNA Accumulation in Nuclear Foci Is Influenced by Viral DNA Replication and Viral Noncoding Polyadenylated Nuclear RNA. Journal of Virology. 92 (13), doi:10.1128/JVI.00220-18, (2018).
AlexaFluor594-5-dUTP | Life Technologies | C1100 | |
anti-DIG FITC | Jackson Lab Immunologicals | 200-092-156 | |
Anti-Rabbit Secondary AlexaFluor594 Monoclonal Antibody | Invitrogen | A-11037 | Goat |
Anti-SSB Antibody | N/A | N/A | Ref. Chiou et al. 2002 |
BLASTn | NIH NCBI | N/A | Free Sequence Alignment Software |
Dextran Sulfate | Sigma Aldrich | D8906 | Molecular Biology Grade |
DIG-Oligonucleotide Tailing Kit | Sigma Roche | #03353583910 | 2nd Gen |
Eight-Chamber Slides | Nunc Lab Tek II | #154453 | Blue seal promotes surface tension but separation by clear gel is also available. |
Formamide | Sigma Aldrich | F9037 | Molecular Biology Grade |
GAPDH Probes | Stellaris | SMF-2019-1 | Compatible with protocol, Quasar 670 |
ImageJ | NIH, Bethesda, MD | N/A | Free Image Analysis Software, [http:rsb.info.nih.gov/ij/] |
OligoAnalyzer | IDT | N/A | Free Oligonucleotide Analyzer |
pcDNA3 | Invitrogen | A-150228 | |
pmaxGFP | Amaxa | VDF-1012 | |
Poly L-Lysine | Sigma Aldrich | P8920 | |
Terminal Transferase | Sigma Roche | #003333574001 | |
Vanadyl Ribonucleoside Complexes | NEB | S1402S | |
Vectashield | Vector Laboratories, Inc. | H-1000 | DAPI within the mounting media scatters the light and reduces contrast. |