Plaquing is a routine method used to quantify live viruses in a population. Though plaquing is frequently taught in various microbiology curricula with bacteria and bacteriophages, plaquing of mammalian viruses is more complex and time-consuming. This protocol describes the procedures that function reliably for regular work with herpes simplex viruses.
There are numerous published protocols for plaquing viruses, including references within primary literature for methodology. However, plaquing viruses can be difficult to perform, requiring focus on its specifications and refinement. It is an incredibly challenging method for new students to master, mainly because it requires meticulous attention to the most minute details. This demonstration of plaquing herpes simplex viruses should help those who have struggled with visualizing the method, especially its nuances, over the years. While this manuscript is based on the same principles of standard plaquing methodology, it differs in that it contains a detailed description of (1) how best to handle host cells to avoid disruption during the process, (2) a more useful viscous medium than agarose to limit the diffusion of virions, and (3) a simple fixation and staining procedure that produces reliably reproducible results. Furthermore, the accompanying video helps demonstrate the finer distinctions in the process, which are frequently missed when instructing others on conducting plaque assays.
The beginnings of virus plaque assays go back to the first discoveries of viruses in the 1890s1. Tobacco mosaic virus was first isolated and passed on tobacco leaves, where individual spots of infection could be recognized and quantified as originating from a single, live virus entity2, later identified as a virion2. Later seminal studies with bacteria and bacteriophages perfected the techniques used to plaque these viruses, including bacteria at the mid-log phase of growth, serial dilution of bacteriophage samples, and top agar with subsequent visualization of literal holes (named plaques) in the bacterial lawn3.
Plaquing of animal viruses lagged the exciting research being conducted with bacteriophages, mainly because the methods required for growing mammalian cells in culture were not developed until the 1940s4. However, the advent of growing murine cells in the absence of the entire host organism4 spawned a new era in the ability to culture and count viruses. Such work was extended for the propagation and quantitation of Western Equine Encephalomyelitis virus in chicken cells and poliovirus in human cells5,6. As the realm of culturable mammalian cells expanded, the bevy of different host cells for various viral infections gave the world a cornucopia of possibilities to study all manner of viruses7. This included the propagation and quantification of human herpesviruses, particularly herpes simplex virus-1 (HSV-1) and -2 (HSV-2), which cause mucocutaneous lesions8. Importantly, all plaque assays are dependent on the existence of live virions, which can enter host cells in a receptor-mediated fashion in a sample9. Regardless of the ubiquity and multitude of publications on the execution of plaque assays5,10,11,12,13,14,15,16, these methods for HSV-1/-2 are a mixture of both art and science; one cannot conduct the assay without proper attention to every detail in the protocol, nor can one execute a successful assay without a crticial eye for subtlety in the process. This manuscript depicts one of the most consistently reproducible methods for HSV-1/-2 plaque assays, with precise details towards the art of the assay that are seldom discussed.
This current protocol obtains live plaque-forming units (PFU) counts for HSV-1 and -2 reliably. Best results are obtained using Vero cells (transformed African green monkey kidney epithelial cells) at low passage (below passage number 155) and routinely grown in alpha-MEM17 supplemented with 10% fetal calf serum (FCS), L-alanyl-L-glutamine, and an antibiotic/antimycotic mixture18. Vero cells are standardly propagated in this medium two to three times per week at a 1/5 dilution each time.
All procedures with the Vero cells and live herpesviruses have been approved by the Towson University Institutional Biosafety Committee. A generalized scheme of these procedures is represented in Figure 1.
1. Seeding of the Vero cells
2. Sample dilution
3. Addition of virus to the cells
4. Staining for plaques
5. Counting plaques and determining virus titer
Table 1 shows an experiment that has optimal results. All 10-fold dilutions follow an approximately 10-fold decrease in plaque counts. These kinds of data can also be seen in Figure 2, an actual plaque assay where the countable number of plaques fell in the 10-4 range for all three replicates. The same can be seen in Figure 3, the top row, where the countable number of plaques was in the 10-3 dilution.
However, Figure 3, bottom row, shows an entirely different result when a plaque assay is not conducted well. First, there are very few plaques visible from any dilution; hence, the data are unusable. Second, there are visible areas around the top where the Vero cells have entirely lifted from the substrate, evidence either of allowing the cells to dry out in the process or of pipetting too vigorously in that region of the well.
Table 2 also shows a set of potentially aberrant outcomes in plaque counts and requires cautions along the way. Of note, neither the top row nor the bottom row of replicates follows a regular 10-fold difference in plaque counts as one moves from one dilution to the next; these data indicate user error during the virus dilution process. Furthermore, it is not likely that a user would accurately count over 15,000 plaques, so one must call those data into question. Finally, one may note that in replicate 1, 10-6 dilution, there is a plaque count in the 30-300 range; nonetheless, given the poor nature of the rest of the data in this replicate, it is unlikely that the 42 plaque count is an accurate number. Hence, the only believable data in Table 2 are in replicate 2.
Figure 4 shows the same problems with cells drying out or being mishandled during the plaque assay. However, Figure 4 critically shows poor Vero cell seeding; every well shows darker purple spots, indicative of cells not spread well across the well and piled on top of one another in clusters.
Figure 1: Overall plaque assay scheme. Adapted from Biorender template, "Viral Plaque Assay Protocol," by BioRender.com. Please click here to view a larger version of this figure.
Figure 2: Plaque assay for HSV-1 in twelve-well plates. Biological replicates were conducted as outlined in Figure 1 on one sample of HSV-1. Serial 10-fold dilutions of an HSV-1 sample (starting at the 10-2 dilution) were added to the Vero cells, incubated, then stained with crystal violet. Please click here to view a larger version of this figure.
Figure 3: Plaque assay for HSV-2 in six-well plates. Biological replicates were conducted as outlined in Figure 1 on one sample of HSV-2. Serial 10-fold dilutions of an HSV-2 sample (starting at the 10-2 dilution) were added to Vero cells, incubated, then stained with crystal violet. Please click here to view a larger version of this figure.
Figure 4: Plaque assay for HSV-1 in twelve-well plates. Biological replicates were conducted as outlined in Figure 1 on one sample of HSV-1. Serial 10-fold dilutions of an HSV-1 sample (starting at the 10-2 dilution) were added to Vero cells, incubated, then stained with crystal violet. Please click here to view a larger version of this figure.
Dilution factor | ||||
Replicate | 10-3 | 10-4 | 10-5 | 10-6 |
1 | 5721 | 635 | 81 | 5 |
2 | 4592 | 365 | 42 | 0 |
3 | 5,519 | 521 | 53 | 1 |
Table 1: Representative plaque counts from one successful assay. Serial dilutions from each biological replicate were conducted; representative plaque counts (starting at the 10-3 dilution) are shown for each dilution.
Dilution factor | ||||
Replicate | 10-3 | 10-4 | 10-5 | 10-6 |
1 | 15225 | 635 | 450 | 42 |
2 | 4592 | 365 | 42 | 0 |
3 | 5519 | 5400 | 53 | 1 |
Table 2: Representative plaque counts from one unsuccessful assay. Serial dilutions from each biological replicate were conducted; representative plaque counts (starting at the 10-3 dilution) are shown for each dilution.
While plaque assays are almost as old as mammalian cell culture itself, it seems that each lab has its own set of protocols to execute this basic assay5,6,10,11,12,13,14,15,16,20. Although the instructions in each published version of a plaque assay differ ever so slightly, few of these manuscripts elucidate the critical nuances involved in obtaining consistent results.
Hence, there are some parts of this protocol that are more art than science. When Vero cells are seeded in wells for the assay, they must be spread evenly throughout the well (Figure 4). All steps involving a liquid transfer (e.g., steps 3.1, 3.2, 3.5, 3.6, 4.1, and 4.2) must be done carefully to keep the Vero cell monolayer hydrated (Figure 3 and Figure 4). It is also crucial not to touch the bottom of the well with any pipette, or the cells may scrape off, further compromising the final data (Figure 4).
While methylcellulose is advantageous as an overlay material to prevent diffusion of virions across the entire well, alternatives may be used. For example, one may use microcrystalline cellulose21,22 or carboxymethyl cellulose21. Regardless, methylcellulose is one of the easiest and least expensive liquid overlay media to use, though. One may also use a different dye21 or fixing solution (e.g., paraformaldehyde)21 for the staining part of the protocol (step 4), but crystal violet in 50% ethanol is the easiest and least expensive among the options.
While it may be overtly straightforward to conduct these assays, they do require training and practice. It is well worth using samples that are not irreplaceable for learning this method so that one has mastered the technique when it matters most.
Plaque assays themselves should only be used, though, when one needs to count bona fide live virions. Plaquing is a much older methodology to measure virus particles5,10 compared to newer quantitative methods like fluorescence microscopy23, ELISA24, enzymatic activity25,26, or qPCR27. However, these latter methods generate surrogate data that suggest the presence of whole virions, yet may lead to a gross overestimate in viable particles because the user would be quantifying an indicator of infection, not real virions. Fluorescence microscopy measures infectious virions by how many cells exhibit the presence of a protein expressed by the virus but does not necessarily detect complete, replication-competent virions. ELISA similarly detects the presence of virus proteins that may exist without an utterly infectious virion. Assaying enzymatic activity, while suggestive of an intact virion, only shows that a single active enzyme is present and does not reflect the true contagious nature of a whole virion. And qPCR merely quantifies copies of a virus' genome, whether that genome is defective or not. The only method comparable to a plaque assay to quantify live viruses is an endpoint dilution assay28, which still requires a statistical estimation via the Reed-Muench or Spearman-Kärber methods29,30,31 to determine virion concentration.
While plaque assays are a routine method used throughout virology, there are particular cases where they are not the quantitative assay of choice. For example, plaque assays require host cells to attach to a substrate; if the best host cell for infection is non-adherent, plaques will never form32. Some viruses grow more slowly than the typical cell cycle of their hosts, and therefore any burgeoning plaques would be overgrown by the host cells themselves21. Plaques may sometimes not form well, just by the nature of the cytopathic effect, making the determination of bona fide plaques a limitation21.
The approach to plaquing herpesviruses presented here is not necessarily unique. However, this in-depth description of the minor caveats associated with a successful HSV plaque assay may diminish the difficulties experienced by many users, especially novices. Regardless, the methodology described is also broadly applicable to plaque assays with many other viruses, including but not limited to picornaviruses33, orthomyxoviruses34, flaviruses35, coronaviruses36, and many other systems in which a quantifiable amount of infectious animal viruses are required.
The authors have nothing to disclose.
We thank countless students in our labs (PJD and BJM) who have worked with us over the years refining these methods. A special thanks to Stan Person, under whose tutelage this methodology was first developed. This work was partially supported by the Towson University Fisher College of Science and Math Undergraduate Research Support fund and NIGMS Bridges to the Baccalaureate grant 5R25GM058264. This content is solely the authors' responsibility and does not necessarily represent the official views of the National Institutes of Health's National Institute of General Medical Sciences.
12-well plates | Corning | 3512 | |
6-well plates | Corning | 3516 | |
Alpha-MEM | Lonza | 12169F | |
Antibiotic/antimycotic | Gibco | 15240096 | |
Crystal violet | Alfa Aesar | B2193214 | |
DMEM | Gibco | 11965092 | |
Dulbecco's PBS (no Mg++ or Ca++) | Gibco | 14190144 | |
Fetal calf serum | Millipore-Sigma | TMS-013-B | |
L-alanyl-L-glutamine (Glutamax) | Gibco | GS07F161BA | |
Hemacytometer | Thermo Fisher | 02-671-54 | |
Methylcellulose | Millipore-Sigma | 27-441-0 | |
Quaternary agent (Lysol I.C.) | Thermo Fisher | NC9645698 | |
Trypan Blue | Corning | 25900CI | |
Trypsin | Cytiva | SH30042.01 | |
Vero cells | ATCC | CCL-81 |