Here, we present a protocol to infect primary human dermal fibroblast with MCPyV. The protocol includes isolation of dermal fibroblasts, preparation of MCPyV virions, virus infection, immunofluorescence staining, and fluorescence in situ hybridization. This protocol can be extended for characterizing MCPyV-host interactions and discovering other cell types infectable by MCPyV.
Merkel cell polyomavirus (MCPyV) infection can lead to Merkel cell carcinoma (MCC), a highly aggressive form of skin cancer. Mechanistic studies to fully investigate MCPyV molecular biology and oncogenic mechanisms have been hampered by a lack of adequate cell culture models. Here, we describe a set of protocols for performing and detecting MCPyV infection of primary human skin cells. The protocols describe the isolation of human dermal fibroblasts, preparation of recombinant MCPyV virions, and detection of virus infection by both immunofluorescent (IF) staining and in situ DNA-hybridization chain reaction (HCR), which is a highly sensitive fluorescence in situ hybridization (FISH) approach. The protocols herein can be adapted by interested researchers to identify other cell types or cell lines that support MCPyV infection. The described FISH approach could also be adapted for detecting low levels of viral DNAs present in the infected human skin.
Merkel cell polyomavirus (MCPyV) is a small, double-stranded DNA virus that has been associated with a rare but aggressive skin cancer, Merkel cell carcinoma (MCC)1,2. The mortality rate of MCC, around 33%, exceeds that of melanoma3,4. MCPyV has a circular genome of ~5 kb1,5 bisected by a non-coding regulatory region (NCRR) into early and late coding regions1. The NCRR contains the viral origin of replication (Ori) and bidirectional promoters for viral transcription6,7. The early region encodes tumor antigen proteins called large T (LT), small T (sT), 57kT, alternative LT ORF (ALTO), as well as an autoregulatory miRNA1,8,9,10. The late region encodes the capsid proteins VP1 and VP211,12,13. LT and sT are the best-studied MCPyV proteins and have been shown to support the viral DNA replication and MCPyV-induced tumorigenesis5. Clonal integration of MCPyV DNA into the host genome, which has been observed in up to 80% of MCCs, is likely a causal factor for virus-positive tumor development14,15.
The incidence of MCC has tripled over the past twenty years16. Asymptomatic MCPyV infection is also widespread in the general population17,18,19. With the increasing number of MCC diagnoses and the high prevalence of MCPyV infection, there is a need to improve our understanding of the virus and its oncogenic potential. However, many aspects of MCPyV biology and oncogenic mechanisms remain poorly understood20. This is largely because MCPyV replicates poorly in established cell lines11,12,21,22,23 and, until recently, skin cells capable of supporting MCPyV infection had not been discovered22. Mechanistic studies to fully investigate MCPyV and its interaction with host cells have been hampered by a lack of cell culture system for propagating the virus5.
We discovered that primary human dermal fibroblasts (HDFs) isolated from neonatal human foreskin support robust MCPyV infection both in vitro and ex vivo24. From this study, we established the first cell culture infection model for MCPyV24. Building on this model system, we showed that the induction of matrix metalloproteinase (MMP) genes by the WNT/β-catenin signaling pathway and other growth factors stimulates MCPyV infection. Moreover, we found that the FDA-approved MEK antagonist trametinib effectively inhibits MCPyV infection5,25. From these studies, we also established a set of protocols for isolating human dermal fibroblasts24,25, preparing MCPyV virions11,12, performing MCPyV infection on human dermal fibroblasts24,25 and detecting MCPyV proteins by IF staining26. In addition, we adapted the in situ DNA hybridization chain reaction (HCR) technology27 to develop a highly sensitive FISH technique (HCR-DNA FISH) for detecting MCPyV DNA in infected human skin cells. These new methods will be useful for studying the infectious cycle of MCPyV as well as the cellular response to MCPyV infection. The natural host reservoir cells that maintain MCPyV infection and the cells that give rise to MCC tumors remain unknown. The techniques we describe in this manuscript could be applied to examine various types of human cells to identify both the reservoir cells and origin of MCC tumors. Our established methods, such as HCR-DNA FISH, could also be employed in the detection of other DNA tumor viruses and the characterization of host cell interactions.
Human neonatal foreskins were obtained from Penn Skin Disease Research Center. Adult human fibroblasts were obtained from discarded normal skin after surgery. All the protocols were approved by the University of Pennsylvania Institutional Review Board.
1. Isolation of human dermal fibroblasts
2. Recombinant MCPyV virion preparation
3. Infection
4. Immunofluorescent staining
5. In situ DNA-HCR
NOTE: This technique requires that cells be seeded on coverslips. For this purpose, the infection conditions described above (step 3) may be scaled up to the 24-well plate format.
The protocol described in this manuscript allowed isolation of a nearly homogenous population of HDFs (Figure 1). As demonstrated by immunofluorescent staining, almost 100% of the human dermal cells isolated using the conditions described in this protocol were positively stained for dermal fibroblast markers, vimentin, and collagen I24, but negative for human foreskin keratinocyte marker K14 (Figure 1). Figure 2 shows the process of generating MCPyV virions using recombinant MCPyV genome and a virion sample after ultracentrifuge concentration. After visualizing the band of MCPyV virions concentrated in the core of the gradient (marked by an arrow in Figure 2B), 500 µL fractions were collected and MCPyV qPCR was performed to identify the peak fractions. An example of the qPCR analysis data is shown in Figure 2C. In some other experiments, the samples were also analyzed with Western blotting using a rabbit-anti-MCPyV VLP antibody (Supplemental Figure 1). Figure 3 shows immunofluorescent stained images of HDFs infected with MCPyV. To manually quantify positive cells, we counted at least three random views of about 300 cells. We consider cells that are LT positive, VP1 positive, or both LT and VP1 positive to be positive for MCPyV infection. In the experiments shown in Figure 3, over 30% of cells are LT positive and more than 10% are VP1 positive. The MCPyV genomes replicated in the infected cells were detected using both the HCR-DNA FISH (Figure 4A) and Immunofluorescent-HCR-DNA FISH (Figure 4B). While the HCR-DNA FISH reveals the localization of MCPyV DNA present in the replication factory (foci) in Figure 4A, the Immunofluorescent-HCR-DNA FISH methods allows simultaneous detection of both MCPyV DNA and LT protein co-localizing at the replication centers (Figure 4B). The images from the immunofluorescent-HCR-DNA FISH demonstrated that this technique could be used to reveal co-localization of viral DNA and the associated viral protein.
Figure 1. Human dermal fibroblasts isolated from neonatal human foreskin. Human dermal cells cultured in DMEM/F12 medium supplemented with 10% FBS were stained using antibodies against vimentin, collagen I, or keratin 14 (K14). The cells were also counterstained with DAPI. Bar, 50 µm. This figure was adapted from Figure S2 of Liu et al., 201624. Please click here to view a larger version of this figure.
Figure 2. Production of MCPyV virion using recombinant viral genome. (A) A plasmid map of pR17b (MCPyV genome plasmid). (B) A representative picture of an MCPyV virion sample harvested and purified over a gradient. Arrow marks the band of MCPyV virions concentrated in the core of the gradient. (C) The viral genome copy number in each gradient fraction was quantified using qPCR. Core gradient fractions (numbers, counting from the top of the gradient, are indicated at the bottom of the graph). Error bars represent standard error of the mean (S.E.M.) of three technical repeats. Please click here to view a larger version of this figure.
Figure 3. Human dermal fibroblasts support robust MCPyV infection, transcription, and replication. Dermal fibroblasts isolated from human skin were treated with MCPyV virions in DMEM F12 medium containing EGF, bFGF, CHIR99021, and collagenase IV for two days. After changing to fresh DMEM/F12 medium containing 20% FBS for three more days, cells were immuno-stained using the indicated antibodies and counterstained with DAPI. Many of cells not only have highly expressed LT and VP1 but also show robust MCPyV replication foci. This figure was adapted from Figure 3 of Liu et al., 201624. Scale bar, 50 m. Please click here to view a larger version of this figure.
Figure 4. Detection of MCPyV in infected cells using FISH techniques. (A) Human dermal cells cultured in DMEM/F12 containing EGF, bFGF, CHIR99021, and collagenase type IV were treated with MCPyV for 2 days. After changing to fresh DMEM/F12 containing FBS for 3 more days, the cells were subjected to HCR-DNA FISH analysis. The cells were also counterstained with DAPI. (B) Human dermal cells cultured in medium containing EGF, bFGF, and CHIR99021 were either untreated (Mock) or treated (Infected) with MCPyV as described in (A). The cells were subjected to immuno-FISH using LT antibody and MCPyV-specific DNA probes before counterstaining with DAPI. HPV16 probes were used as a negative control. Scale bar, 10 m. This figure was adapted from Figure 4 of Liu et al., 201624. Please click here to view a larger version of this figure.
Probe name | Sequences |
MCPyV probe 1 | CCTCAACCTACCTCCAACTCTCACCATATTCGCTTC TAAAA tgagctacctcactaaggagtggtttttatactgcagtttcccgcccttg ATTTT CACATTTACAGACCTCAACCTACCTCCAACTCTCAC |
MCPyV probe 2 | CCTCAACCTACCTCCAACTCTCACCATATTCGCTTC TAAAA agaggcctcggaggctaggagccccaagcctctgccaacttgaaaaaaaa ATTTT CACATTTACAGACCTCAACCTACCTCCAACTCTCAC |
MCPyV probe 3 | CCTCAACCTACCTCCAACTCTCACCATATTCGCTTC TAAAA cattgactcatttcctggagaggcggagtttgactgataaacaaaacttt ATTTT CACATTTACAGACCTCAACCTACCTCCAACTCTCAC |
MCPyV probe 4 | CCTCAACCTACCTCCAACTCTCACCATATTCGCTTC TAAAA gatactgccttttttgctaattaagcctcttaagcctcagagtcctctct ATTTT CACATTTACAGACCTCAACCTACCTCCAACTCTCAC |
MCPyV probe 5 | CCTCAACCTACCTCCAACTCTCACCATATTCGCTTC TAAAA aagcttctcctgtaagaatagcttccaaagttactcctgtggtggcactt ATTTT CACATTTACAGACCTCAACCTACCTCCAACTCTCAC |
MCPyV probe 6 | CCTCAACCTACCTCCAACTCTCACCATATTCGCTTC TAAAA ggatgttgccataacaattaggagcaatctccaaaagcttgcacagagcc ATTTT CACATTTACAGACCTCAACCTACCTCCAACTCTCAC |
MCPyV probe 7 | CCTCAACCTACCTCCAACTCTCACCATATTCGCTTC TAAAA gctcaggggaggaaagtgattcatcgcagaagagatcctcccaggtgcca ATTTT CACATTTACAGACCTCAACCTACCTCCAACTCTCAC |
MCPyV probe 8 | CCTCAACCTACCTCCAACTCTCACCATATTCGCTTC TAAAA aagcctgggacgctgagaaggacccatacccagaggaagagctctggctg ATTTT CACATTTACAGACCTCAACCTACCTCCAACTCTCAC |
MCPyV probe 9 | CCTCAACCTACCTCCAACTCTCACCATATTCGCTTC TAAAA agcttcgggaccccccaaattttcgctttcttgagaatggaggaggggtc ATTTT CACATTTACAGACCTCAACCTACCTCCAACTCTCAC |
MCPyV probe 10 | CCTCAACCTACCTCCAACTCTCACCATATTCGCTTC TAAAA cttttggctagaacagtgtctgcggcttgttggcaaatggttttctgaga ATTTT CACATTTACAGACCTCAACCTACCTCCAACTCTCAC |
HPV probe 1 | CCTCAACCTACCTCCAACTCTCACCATATTCGCTTC TAAAA cagctctgtgcataactgtggtaactttctgggtcgctcctgtgggtcct ATTTT CACATTTACAGACCTCAACCTACCTCCAACTCTCAC |
HPV probe 2 | CCTCAACCTACCTCCAACTCTCACCATATTCGCTTC TAAAA acaatattgtaatgggctctgtccggttctgcttgtccagctggaccatc ATTTT CACATTTACAGACCTCAACCTACCTCCAACTCTCAC |
HPV probe 3 | CCTCAACCTACCTCCAACTCTCACCATATTCGCTTC TAAAA gtcagctatactgggtgtaagtccaaatgcagcaatacaccaatcgcaac ATTTT CACATTTACAGACCTCAACCTACCTCCAACTCTCAC |
HPV probe 4 | CCTCAACCTACCTCCAACTCTCACCATATTCGCTTC TAAAA ctttggtatgggtcgcggcggggtggttggccaagtgctgcctaataatt ATTTT CACATTTACAGACCTCAACCTACCTCCAACTCTCAC |
HPV probe 5 | CCTCAACCTACCTCCAACTCTCACCATATTCGCTTC TAAAA ccatccattacatcccgtaccctcttccccattggtacctgcaggatcag ATTTT CACATTTACAGACCTCAACCTACCTCCAACTCTCAC |
Table 1: Probes used in the study.
Supplemental Figure 1. Screening Iodixanol gradient fractions for MCPyV VP1. MCPyV-infected cells were lysed and subjected to Iodixanol gradient fractionation. Core gradient fractions (numbers, counting from the bottom of the gradient in this experiment, are indicated at the top of the figure) were subjected to Western blot analysis to detect VP1. 10 µL samples of each fraction were adjusted to 1x load dye with a 5% final concentration of 2-mercaptoethanol and briefly heated to 65 °C. The samples were separated on a 4-12% bis-Tris gel and blotted onto nitrocellulose. Western blotting was conducted in TBST with nonfat dry milk using a rabbit-anti-MCPyV VLP antiserum diluted 1:5000. The antiserum is available upon request. The 50 kDa standard (which migrates similarly to the 47 kDa MCPyV VP1 monomer) is marked with an asterisk. In the shown image, sample reduction was incomplete and disulfide-linked VP1 multimers are apparent. Use of dithiothreitol and/or higher denaturation temperatures can result in more complete reduction to the VP1 monomer species. Please click here to view a larger version of this figure.
The methods described above , including isolation of dermal fibroblasts from human skin tissue, preparation of recombinant MCPyV virions, infection of cultured cells, immunofluorescent staining, and a sensitive FISH method adapted from HCR technology, which should enable researchers to analyze MCPyV infection27. One of the most critical steps to achieving MCPyV infection in vitro is the production of high-titer virion preparations. Using the protocol for preparation of recombinant MCPyV virions described here, we achieve viral titers of 1012 genome copies per mL of idoxanol solvent11,12. These high-titer MCPyV preparations allowed us to identify HDFs as the only primary skin cell type thus far that appears to support the full MCPyV infection cycle24.
Isolating fresh human dermal fibroblasts using the protocol described in this manuscript has enabled us to control the quality of the cells used in MCPyV infection experiment and ensure consistency of findings across cells isolated from different patient samples. All human dermal fibroblasts isolated from neonatal skin should be assessed for presence of MCPyV DNA by qPCR to ensure the absence of MCPyV infection prior to treatment with recombinant MCPyV virions. In order to obtain the maximum MCPyV infection efficiency, it is critical to use human dermal fibroblasts isolated freshly from neonatal human foreskin as higher passage fibroblasts support lower MCPyV infection efficiencies. Further, when isolated in parallel from adult skin samples, fibroblasts from neonatal skin samples supported much higher MCPyV infection efficiency24. We have never tested any commercially available cells for MCPyV infection. However, we see no specific reason that commercially isolated fibroblasts should be inferior other than if they are higher passage at the point of freezing.
We have also used the protocol described here to isolate dermal fibroblasts from other animals. For example, we have used the protocol to successfully isolate dermal fibroblasts from mouse (Mus musculus), rat (Rattus norvegicus), tree shrew (Tupaia Belangeri), and rhesus macaque (Macaca mulatta)25. As observed in human cells, fibroblasts isolated from younger chimpanzees allowed much more efficient MCPyV infection compared to the older animals25.
Compared to traditional FISH, HCR-DNA FISH is a simple, but highly sensitive method for detecting viral genomes24,27. Centers of MCPyV replication can be readily detected in the nuclei of infected cells as highly specific punctate foci, with little to no background signal detected in the same samples treated with a HPV specific probe (Figure 4)24. In the future, the approach could therefore be explored to detect low-abundance MCPyV DNA present in infected human skin. It could also be adapted to monitor other DNA viruses. When combined with immunofluorescent staining, the immuno-FISH provides a powerful method for simultaneously detecting viral DNA and either viral encoded protein or host proteins present in the same cells (Figure 4). For obtaining the optimum results using the protocols described here, it is best to use freshly fixed samples for immunofluorescent staining and HCR-DNA FISH analysis. For the HCR-DNA FISH analysis, the probes and amplifier should be stored in -80 °C in small aliquots to avoid repeated freezing and thawing.
Using the HDF cell culture and MCPyV infection protocol, we explored the cell growth conditions that best support MCPyV entry, transcription, and replication in HDFs. We discovered that treatment of HDFs with growth factors, such as EGF, bFGF, and stimulation of the WNT signaling pathway significantly promote MCPyV infection. Gene expression profiling reveals that, upon stimulation with these growth factors and activation of the WNT signaling pathway, several genes of the matrix metalloproteinase (MMP) family, including MMP1, 3, 7, 9, 10, 11 and 13, were robustly induced. Because these MMP enzymes are capable of degrading extracellular matrix proteins, we reason that they may stimulate MCPyV infection by disrupting the extracellular matrix of the host cells. Indeed, treating HDFs with collagenase type IV, a member of the MMP family, robustly stimulates MCPyV infection (Figure 3)24. We also screened an array of compounds, including several FDA-approved kinase inhibitors, for inhibitory effect on MCPyV infection. We discovered that trametinib, a MEK1 and MEK2 inhibitor, dramatically inhibits MCPyV infection24. Others may use or modify this infection system as a drug discovery platform for MCPyV inhibitors that reduce the MCPyV viral load in immunocompromised patients.
In summary, these protocols for achieving highly efficient MCPyV infection provide a platform to elucidate the poorly understood MCPyV infectious cycle and MCPyV-induced oncogenic mechanisms in the context of viral infection. This set of protocols could be applied in future studies to explore additional MCPyV-permissive human cell types, and the original cells of MCC, in which incidental MCPyV infection could lead to tumor development.
The authors have nothing to disclose.
The authors would like to thank Dr. Meenhard Herlyn (Wistar Institute) and Dr. M. Celeste Simon (University of Pennsylvania) for providing reagents and technical support. We also thank the members of our laboratories for helpful discussion. This work was supported by the National Institutes of Health (NIH) grants (R01CA187718, R01CA148768 and R01CA142723), the NCI Cancer Center Support Grant (NCI P30 CA016520), and the Penn CFAR award (P30 AI 045008).
Fetal calf serum | HyClone | SH30071.03 | |
MEM Non-Essential Amino Acids Solution, 100X | Thermo Fisher Scientific | 11140050 | |
GLUTAMAX I, 100X | Thermo Fisher Scientific | 35050061 | L-Glutamine |
DPBS, no calcium, no magnesium | Thermo Fisher Scientific | 14190136 | |
0.05% Trypsin-EDTA | Thermo Fisher Scientific | 25300-054 | |
DMEM/F12 medium | Thermo Fisher Scientific | 11330-032 | |
Recombinant Human EGF Protein, CF | R&D systems | 236-EG-200 | Store at -80 degree celsius |
CHIR99021 | Cayman Chemical | 13122 | Store at -80 degree celsius |
CHIR99021 | Sigma | SML1046 | Store at -80 degree celsius |
Collagenase type IV | Thermo Fisher Scientific | 17104019 | |
Dispase II | Roche | 4942078001 | |
Antibiotic-Antimycotic | Thermo Fisher Scientific | 15240-062 | Protect from light |
DMEM medium | Thermo Fisher Scientific | 11965084 | |
Alexa Fluor 594 goat anti-mouse IgG | Thermo Fisher Scientific | A11032 | Protect from light |
Alexa Fluor 488 goat anti-rabbit IgG | Thermo Fisher Scientific | A11034 | Protect from light |
OptiPrep Density Gradient Medium | Sigma | D1556 | Protect from light |
Paraformaldehyde | Sigma | P6148 | |
anti-MCPyV LT (CM2B4) | Santa Cruz | sc-136172 | Lot # B2717 |
MCV VP1 rabbit | Rabbit polyclonal serum #10965 | https://home.ccr.cancer.gov/lco/BuckLabAntibodies.htm | |
Hygromycin | Roche | 10843555001 | |
Basic Fibroblast Growth Factors (bFGF), Human Recombinant | Corning | 354060 | Store at -80 degree celsius |
Benzonase Nuclease | Sigma | E8263 | |
Plasmid-Safe ATP-Dependent DNase | EPICENTRE | E3101K | |
Probe hybridization buffer | Molecular technologies | ||
Probe wash buffer | Molecular technologies | ||
Amplification buffer | Molecular technologies | ||
Alexa 594-labeled hairpins | Molecular technologies | B4 | Protect from light |
Triton X-100 | Sigma | X100 | |
Quant-iT PicoGreen dsDNA Reagent | Thermo Fisher Scientific | P7581 | |
BamHI-HF | NEB | R3136 | |
Buffer PB | Qiagen | 19066 | |
blue miniprep spin column | Qiagen | 27104 | |
50mL Conical Centrifuge Tubes | Corning | 352070 | |
T4 ligase | NEB | M0202T | |
MagicMark XP | Thermo Fisher Scientific | LC5602 |