key: cord-0733968-bl70k152
authors: Grigorov, Boyan; Rabilloud, Jessica; Lawrence, Philip; Gerlier, Denis
title: Rapid Titration of Measles and Other Viruses: Optimization with Determination of Replication Cycle Length
date: 2011-09-07
journal: PLoS One
DOI: 10.1371/journal.pone.0024135
sha: 0b67e259fd64cd00d911f3f78b16e5c9c5fa04ab
doc_id: 733968
cord_uid: bl70k152

BACKGROUND: Measles virus (MV) is a member of the Paramyxoviridae family and an important human pathogen causing strong immunosuppression in affected individuals and a considerable number of deaths worldwide. Currently, measles is a re-emerging disease in developed countries. MV is usually quantified in infectious units as determined by limiting dilution and counting of plaque forming unit either directly (PFU method) or indirectly from random distribution in microwells (TCID50 method). Both methods are time-consuming (up to several days), cumbersome and, in the case of the PFU assay, possibly operator dependent. METHODS/FINDINGS: A rapid, optimized, accurate, and reliable technique for titration of measles virus was developed based on the detection of virus infected cells by flow cytometry, single round of infection and titer calculation according to the Poisson's law. The kinetics follow up of the number of infected cells after infection with serial dilutions of a virus allowed estimation of the duration of the replication cycle, and consequently, the optimal infection time. The assay was set up to quantify measles virus, vesicular stomatitis virus (VSV), and human immunodeficiency virus type 1 (HIV-1) using antibody labeling of viral glycoprotein, virus encoded fluorescent reporter protein and an inducible fluorescent-reporter cell line, respectively. CONCLUSION: Overall, performing the assay takes only 24–30 hours for MV strains, 12 hours for VSV, and 52 hours for HIV-1. The step-by-step procedure we have set up can be, in principle, applicable to accurately quantify any virus including lentiviral vectors, provided that a virus encoded gene product can be detected by flow cytometry.

Determining the amount of infectious virus is a crucial issue for any virologist. To date, different methods have been used for viral titration depending on the virus concerned. The most popular ones are the plaque forming units (PFU) [1] and the 50% tissue culture infective dose (TCID50) [2] . They are used for cytopathic viruses (e.g. HIV-1, poliovirus, Japanese encephalitis virus, measles virus, etc…) and are based on serial dilutions of the viruscontaining samples and observation of the appearance of a cytopathic effect (CPE) in a cell monolayer. The PFU technique measures the number of virus particles capable of forming plaques per volume unit, and accounts for the replication potency of the virions, i.e. it is a functional measure. The PFU assay is laborious, poorly automatable, and requires several hours to several days according to the speed of virus growth and propagation. It also suffers from subjectivity because of the tedious manual plaque counting and possible plaque-like defects in the cell monolayer. The TCID50 method is a statistical derivative of the PFU assay.

Instead of counting individual plaques, multiple replicates of each virus dilution are made and the TCID50 titer is calculated from the 50% endpoint where half of the replicates contained at least one PFU. Wells with destroyed cell monolayers are easily counted either manually under the microscope or automatically using a viable colorimetric assay. TCID50 values are intrinsically discontinuous, and the value coverage is non-homogeneous (i.e. intervals between two discontinuous values are not identical) as illustrated in Figure 1 . Therefore, increasing the measurement accuracy would require a number of replicates high enough to discourage routine usage. Handling and time requirements of TCID50 and PFU methods are comparable. With both assays, virus titration is more difficult with cells growing in suspension.

Due to the inconveniences mentioned above for these two titration methods, novel, quicker and less cumbersome techniques have been developed. Such titration procedures have been developed for influenza viruses, adenoviruses, HIV-1, SV40, human coronaviruses, hepatitis A virus, as well as for recombinant and/or virus-like particles [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] . Many of these techniques are based on the identification of infected cells at the single cell level using immunodetection of viral proteins (structural or glycoproteins) by flow cytometry [9] , [7] , [13] [14] . In such studies, the virus titer is calculated from the proportion of infected cells (i.e. positively labeled) after exposure of a given number of indicator cells to a given virus suspension volume. In each case, the assumption is made that one cell is infected by a single infectious virus particle. However, at a too high multiplicity of infection (MOI) (i.e. the ratio of infectious virus particle per cell) such an assumption becomes erroneous, leading to an underestimation of virus titers. Indeed, the higher the MOI, the higher the probability of multiple viruses infecting a single cell. Likewise, secondary infections of cells by newly produced viruses during the assay lead to overestimation of the virus titer.

To avoid these pitfalls, we have explored the prerequisites and limitations of the use of flow cytometry for virus quantification with a specific focus on measles virus (MV) strains, while taking into account the following parameters: (i) virus infected cells must be clearly distinguished from the uninfected ones by emitting a fluorescent signal above the autofluorescence background; (ii) the virus titer is best determined by averaging titers calculated from several amounts of the viral inoculum; (iii) the measurement should be optimally done at, or just prior to, the completion of the first viral replication cycle to avoid secondary infection; (iv) virus titer should be calculated from the number of uninfected cells by applying the Poisson probability distribution and (v) the method should be validated by comparison with an already existing titration technique e.g. TCID50. In addition, the assay could be used to approximate the duration of a single replication cycle. Our method was validated by the strong correlation found with the TCID50 technique over a large set of samples. The assay lasts less than two days which favorably compares with the ,1-2 week duration of the conventional PFU or TCID50 methods currently used for MV titration.

In order to validate this rapid assay for measles virus titration, we have compared it to the classical TCID50 method. A high amount of virus samples quantified by both techniques showed a strong correlation indicating thus the accuracy of the novel method. The detection limit is 10 4 infectious units/ml.

To illustrate its wide application, our method was also adapted to titrate recombinant MV and vesicular stomatitis virus (VSV) expressing the green fluorescent protein (GFP) as gene reporter of the infection. Likewise, the method was also applicable to measure HIV-1 titer using a reporter cell line that conditionally expresses GFP under the control of the Tat-inducible HIV-1 LTR promoter. In the latter case, titer accuracy required the blocking of cell growth by the DNA polymerase inhibitor aphidicolin over the 2-day measurement period.

We have used TCID50 titration method as a reference for the validation of our rapid titration technique. TCID50 is based on the end-point dilution of the virus at which a cytopathic effect (CPE) is detected in 50% of the cell culture replicates infected by a given amount of virus suspension [2] . The latter was serially diluted in 0.25 ml of DMEM medium supplemented with 6% fetal calf serum (FCS) and antibiotics (culture medium) by serial transfer of 0.027 ml (i.e. ten-fold dilution) in the first row of a 96 well plate. When diluting the virus suspension, each micropipette tip ought to be changed at every dilution to avoid uneven distribution of Figure 1 . Virus titration by TCID50 is intrinsically discontinuous. Discontinuous titer values given by all possible combinations of percentage of infected wells above, at, and below 50% of eight replicates within one Log range. Note that (i) some mantissa are identical for two or four combinations (indicated in italics in x axis) and (ii) identical sets of mantissa of the logarithm are obtained at any place below and above the displayed titer range. Vertical bars indicate the variable interval separating one titer value from its nearest superior and inferior titer value. doi:10.1371/journal.pone.0024135.g001 infectious particles that remain adsorbed on the tip. Subsequently, 0.030 ml of each dilution was distributed vertically, i.e. 8 replicates for each dilution. Permissive cells, i.e. Vero cells, (10,000 cells in 0.2 ml) were seeded in each well, and incubated at 37uC in 5% C0 2 and humid atmosphere. CPE induced by measles virus could be observed under the microscope after 3-4 days, but because CPE located in the well edge could hardly be spotted, the final counting of wells with CPE was done after 10 days. This incubation time can be much shorter for quicker growing viruses (i.e. 1-2 days for VSV). To avoid underestimation of virus titer, the incubation time should be adapted to the virus species and physicochemical conditions used for virus propagation. TCID50 was then calculated using the formula:

where ''Above 50%'' dilution is the last dilution expressed as the denominator of 1 (i.e. 250 for 1/250 dilution) for which $50% wells display CPE, (''Above 50%'' and ''Below 50%'' values are fraction of wells with CPE at the last dilution for which $50% wells display CPE and fraction of wells with CPE at the next dilution, respectively, and v the virus volume (in ml) from the undiluted sample used to inoculate the wells.

Step-by-step MV titration procedure by flow cytometry (protocol at a glance)

The flow cytometry titration method, based on the immunodetection of the viral F-glycoprotein on the surface of infected cells, was performed as shown below (an example of an assay timing is indicated when taking into account the following parameters: (i) titration of ,20 samples of an MV strain with (ii) a replication cycle of 18 hours, (iii) without interruption of the assay until the calculation of the titer is done using the online spreadsheet).

1. Start -cell seeding (10h00). Seed 100 000 of Vero cells for MV laboratory/vaccine strains or Vero-Slam cells for MVwild type strains in 0.25 ml of culture medium in each well of a 48 well plate. Preferably, plate the cells in the morning and infect them in the afternoon. Alternatively, inoculate the cells just after seeding. In that case perform the seed and inoculation in the afternoon. Do not seed the cells the day prior to the inoculation because of resulting change in cell number due to cell growth. The cell number seeded should be carefully adjusted since it is part of the formulae to calculate the virus titer in IU/ml (see below). Likewise, during the assay, cell growth should be very limited hence cell plating and infection on the same day. Indeed any overgrowth of uninfected cells versus infected cells during the titer assay will result in the underestimation of the virus titer (see also the application for titration of HIV-1 below).

2. Virus inoculation (16h00). Prepare and add three-fold dilutions of the virus suspension to the cells in duplicate (see Figure 2 for a ''protocol at a glance''). Fix the inoculation volume Day+1 12h00 (lunch break 12h00-14h00) N At this stage, the cells could be kept for 1 or 2 days at 4uC in the dark prior to the flow cytometry analysis.

! Note that controls should include uninfected cells that have been treated according to the abovementioned procedure! Option 2 -for MV wild type strains. Perform cytometry 24 hours post inoculation following the same steps as described in ''option 1''.

Option 3 -for MV GFP encoding viruses. Perform all the steps described in ''option 1'' except primary and secondary antibody labeling. After trypsinization, fix cells and perform the flow cytometry analysis.

5. Flow cytometer acquisition (Day+1 14h00). Set the side scatter (SSC) and the forward scatter (FSC) to select viable cells on unlabeled cell sample. Adjust the fluorescence background of the unlabeled cell sample to the lowest intensity (i.e. to have the fluorescent histogram close to the ordinate). Set the fluorescent threshold for positivity by gating more than 99.5% of the uninfected control cells outside the gate. Analyze every virus dilution by flow cytometry.

6. Titer calculation (End of the titration assay Day+1 16h00). First, determine the number of non-infected cells after infection, and then calculate for each sample the number of noninfected cells (N[n.i.]) within the 100 000 cell input according to the following N n:i: ½ ~100,000

where F + cells are cells with fluorescence above the .99.5% threshold defined above. Example: For a threshold set to 0.3 %F + on the uninfected control sample, and a percentage of infected cells %F + = 10.3% in a test inoculated sample, then the number of non-infected after virus inoculation is

Calculate the titer from the number of uninfected cells: this titer estimation is based on the fact that the probability P(n) that one cell could be infected by ''n'' viral particles follows the Poisson law of the parameter ''l'' which is equal to the multiplicity of infection (MOI).

The probability that one cell will not be infected is then equal to P(0) = e 2l . P(0) could be calculated experimentally by determining the proportion of uninfected cells from the total number of cells: The Poisson's law is valid provided that the total number of cells is higher than 10, the number of infectious virus is higher than 30 and the MOI is below 10. While the two former parameters are always respected because of the high cell number and minimal virus input, MOI could be well over 10 for virus-rich samples. When the percentage of infected cells exceeds 30-40% the titer value tends to be underestimated. Thus, exclude values .40% infected cells from the calculation Likewise, exclude values below ,0.5% infected cells since they fell too close to the background limit to be taken into account. Average the titers calculated from consecutive 3-fold dilution ranges showing ,3-fold intervals.

For automatic titer calculation, an online spreadsheet (csv file) is provided where the results from the flow cytometry for all viral inputs are simply entered where indicated. This file can be downloaded by following the link below (Ctrl+click): https://spreadsheets.google.com/ccc?key = 0AvRhxJMlLBAAdH-VaUk1fMDktOXZYQnA3dUo4UEgwV3c&hl = en You can also copy and paste the link into your browser. Then, go to ''File''RdownloadRExcel

Cells: Vero -ATCC CCL-81; Vero-SLAM [15] ; GHOST Cell Transfectants -GHOST (3) 

In order to validate the rapid protocol for measles virus titration, we have compared it to the classical TCID50 method. The latter was chosen instead of the PFU method since it is less laborious and operator dependent, there are no plaque-counting difficulties [18] , and it has been routinely used by our laboratory over 20 years after internal validation in agreement with a large-scale European assessment [19] and WHO recommendations (http://www.who.int/ vaccines-documents/DocsPDF-IBI-e/mod7_e.pdf ). Six independent titrations of a given viral stock of MV-Schwarz strain gave a similar titer with 1% variation compared to 1.5% using the TCID50 titration method for the same sample. This demonstrates that the flow cytometry-based titration method is reproducible. Furthermore, titers of near one hundred MV stocks measured by both flow cytometry and TCID50 assay showed a high level of correlation (Fig. 3 , r = 0.930, 2a,0.001).

After elaborating the novel titration assay, we defined its detection limit. A series of 10-fold dilutions of an MV-Schwarz strain stock (6.3 Log IU/ml) was used to evaluate the sensitivity of the technique. To allow the increase of the inoculation volume and to compensate for low viral concentration of some samples, 12-well plates and 4610 5 cells were used. Virus concentration, smaller than 4 Log IU/ml, could not be accurately measured by the flow cytometry assay, even when using higher inoculation volumes than 0.3 ml in a final volume of 1 ml of growth medium.

Wild type MV strains are routinely isolated from infected patients by growing them exclusively on CD150/SLAM expressing cells. In the past, other MV strains have been isolated, passed on Vero cells with acquisition of a tropism for CD46 due to only a few point mutations in the H protein, the most prominent of which is N481Y [20] . These strains are called ''laboratory strains'' (or Edmonston-like). Some of them have been further attenuated by forced growth in chicken embryonic fibroblasts to give rise to the currently widely used measles vaccine strains. Wild type MV and laboratory/vaccine strains are titrated using Vero-CD150/SLAM, and Vero cells, respectively. 

Not all viruses produced by the target cells are infectious. Indeed, a large proportion of the virus progeny is defective (noninfectious), and could bind to, and possibly enter into, host cells. Consequently, viral glycoproteins may be found at the cell surface due to the fusion of the viral and cellular membranes leading to an overestimation of the number of infected cells.

To rule out this hypothesis, the rapid titration method was applied to an MV suspension inactivated by UV irradiation at 254 nm for 25 min. Twenty hours post infection and even at the highest virus input (equivalent to MOI = 10), no F protein labeling was detected from the UV inactivated virus while the nonirradiated control virus stock showed the expected titer value of 10 6 IU/ml (data not shown). Similar results were obtained when MV was heated at 60uC for 20 min, a condition that abolishes virus fusion but not virus binding to the cellular receptor [21] . This indicates that defective virions are not taken into account in the virus titration based on flow cytometry, likely because of the internalization of the adsorbed viral glycoproteins in agreement with the endocytosis of MV bound to its cellular receptor [22] .

Thus, the flow cytometry based rapid method measures infectious virus particles (viruses that can undergo at least one complete infection cycle). The titration technique by flow cytometry is based on a single round of infection and can be used to determine the duration of the virus replication cycle

The major pitfall that could compromise the accuracy of virus titration by flow cytometry is the overestimation linked to data poisoning by secondary infection. Indeed, during the assay, infected cells progressively become producers of new virions that may spread to cells that have initially escaped infection by the inoculum. Therefore, it is essential to determine the optimal interval separating the virus inoculation and the cell harvest for flow cytometry analysis. surface of an infected cell. When the cells were infected with smaller inoculums, the histograms of infected cells were progressively and regularly shifted to the left. This held true until 18 h.p.i. At 24 h.p.i., the inoculation with the second largest volume (0.010 ml) resulted in a peak shift that partially overlapped with that observed with 0.030 ml (Fig. 5 , compare pink and green histograms). At the last time point studied (48 h.p.i.), histograms from the first five virus inputs almost completely overlapped, and even a very small inoculum (MOI as low as 10 24 ) gave rise to a significant percentage of infected cells (black histogram). The time interval allowing maximal expression of the viral gene with the highest inoculum, but not allowing similar expression levels with a lower inoculum size was thus considered to represent the duration of a single virus replication.

For the wild type MV-G954 strain, the optimal incubation time was slightly delayed to 24 h. (Fig. 5, panel ''B'') .

To further confirm the above observations, two complementary experiments using entry inhibitors for de novo produced virions were carried out. First, we took advantage of a fusion inhibitory peptide -FIP (z-D-Phe-L-Phe-L-Gly), which above 0.1 mg/ml fully prevents MV entry (Fig. 6A, secondary infection by MV-eGFP virus. While such treatment almost completely abolished infection, the delayed addition of the antiserum by 6 hours did not interfere with the virus titration (Fig. 6) . This strongly suggests that the time interval we have graphically selected as described in Fig. 5 is close to a single virus infection cycle and is optimal to quantify any given infectious virus sample.

Thus, provided that the duration of infection allows only a single cycle of virus replication, the rapid method for measles virus titration is accurate.

The optimal time to titer another virus, VSV-gfp, was determined as 8 h.p.i. (Fig. 5 panel ''C'') , which fit well with the 6-8 hours duration of the VSV replication cycle [23] , [24] . When the apparent titers were blindly calculated for each inoculum and for every time interval, their plotting against the inoculum size gave rise to an almost perfect and expected horizontal distribution (i.e. an identical titer value independent of inoculum size) only at the optimal time for each virus (i.e. 18 h.p.i., 24 h.p.i. and 8 h.p.i. red curves, bottom graphs). Finally, the titer, determined by the rapid titration procedure at 8 h.p.i. correlated best with the titer obtained by the classical TCID50 method.

The protocol was adapted to quantify HIV-1 using indicator cells expressing a reporter gene product upon activation from an incoming virus. GHOST HIV-1 indicator cells encode for a GFP protein under the control of HIV-1 LTR and thus respond to Tat protein expression resulting from an HIV-1 infection by becoming (Fig. 7A) [25] . As HIV-1 infection is rather slow, a kinetics experiment was done over 72 hours. To overcome any interference of cell number changes due to proliferation during the titration assay, the experiment was duplicated in the presence of aphidicolin, a reversible inhibitor of the DNA polymerase. Indeed, after one day lag period, cells started to proliferate to double in number within two days of infection (Fig. 7B) and stop growing thereafter. The aphidicolin treatment prevented any increase in cell number without affecting cell viability over the first 48 h. Virus titration over a wide range of inoculation volumes was found to be more reliable in the presence than in the absence of aphidicolin (compare Fig. 7C with Fig. 7D online) , and titers determined after 48 h of culture were within the error range of that determined by the TCID50 assay.

Different virus inputs (10 ml, 3 ml, 1 ml, 0.3 ml) from a concentrated stock of HIV-1 based lentiviral particles (pseudotyped with VSV envelope G protein and expressing GFP as a transgene [26] [27] [28] ) were used to infect Vero cells according to the procedure already described above. 24 hours afterwards, the cells were trypsinized and GFP positive cells were measured by flow cytometry. The titer was calculated for each input, and values below 50% of F + cells were averaged to calculate the titer. Example: for virus volumes 0.3 ml, 1 ml, 3 ml and 10 ml the percentage of infected cells was 15.3%, 43.7%, 72.4% and 96.3%, respectively. Therefore the virus titer (as average) was calculated from the 2 lowest flow cytometry values and was 7.75 Log IU/ml (which corresponds to (7.74 Log IU/ml for 15.3%+7.76 Log IU/ml for 43.7%)/2 = 7.75).

In comparison with other flow-cytometry based methods described previously to titer different viruses [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] , our technique appears original and complementary in several aspects: (i) a complete set of critical parameters has been fully explored and experimentally validated; (ii) our method is the first one to describe a flow-cytometry based procedure to determine the duration of a single virus replication cycle, a key parameter for titration accuracy, and (iii) the first rapid method for quantifying different measles virus strains; (iv) adjustments of critical parameters allowed its use to titer two other unrelated viruses, VSV and HIV-1; (v) by applying the Poisson law the titer calculation appeared to be very precise and highly reproducible.

Compared to the PFU method and TCID50, the flow cytometry based method has the advantage of being rapid which might be crucial in vaccine production and when short deadlines in research laboratories should be met. Moreover, it could be used to titer non-replicative virus particles (e.g. lentiviral vectors), and a large number of virus samples may be compared simultaneously thanks to limited handling and high precision. However, there are two major limitations: (i) the detection limit is 10 4 IU/ml, which prevents its use for titration of clinical samples that often have a low virus load, and (ii) it is not recommended for viruses (mutants, etc.) with replication cycles exceeding more than 2 days. Alternative methods have been designed based on the measure of either the genome content of a viral suspension [12] or cellassociated viral nucleic acids by quantitative real-time PCR after end point dilution of the infecting virus (PCR infectivity assay) [29] [30] . However, the obtained viral nucleic acid values have to be calibrated using an infectious virus reference, and any change in the [nucleic acid/infectious unit] ratio prevents the accurate determination of the number of infectious particles. The advantages/disadvantages and specific required equipment of every titration methods have been listed in Table 2 as a helpful tool to choose the most suitable titration procedure.

The virus titration method based on the detection of infected cells by flow cytometry after a time interval corresponding to one virus replication cycle proved to be reproducible, accurate, rapid (one virus replication length), and with a detection limit of 10 4 IU/ ml. Furthermore, it could be easily automated [31] . Titers obtained through this technique correlate significantly with those determined by the classical TCID50 method but with higher accuracy and a 3-fold lower range of confidence limit. Such a method may be particularly beneficial for studies requiring titration of large numbers of MV stocks and especially when strict deadlines should be respected. In principle this method can be adapted to titer any virus for which a viral protein can be easily detected by flow cytometry, including recombinant viruses carrying a fluorescent reporter gene (or any gene encoding a cell surface protein that could be immunolabelled) and the use of indicator cells with an endogenous reporter gene that is inducible upon virus infection (see Table 1 for setting up steps of a new virus titration assay). The method is not suitable for titration of viruses (e.g. recombinant viruses, mutants, etc.) which replication cycle takes more than 30 h unless the cell proliferation is prevented. Indeed, the accuracy of virus titration critically relies on the precision of seeded cell number that is used in the formulae to calculate the titer. Any significant cell growth during the assay may result in the underestimation virus titer because infected cells usually grow slower than uninfected cells. This method could prove useful to titrate HIV and HIV molecular clones designed to undergo a single viral cycle. Moreover, this method also allows determination of the duration of a virus replication cycle, a key parameter for the accuracy of the virus titration. 3. Perform titration of your viral stock using an already described and validated method, i.e. TCID50. Note the titre. 4. Define the single replication cycle by analyzing the flow cytometry histograms: note the time point where the peak of the second virus input starts to approach and overlap the peak of the first (highest) virus input) 5. Correlate the data from point 2, 3 and 4 above. The duration of the replication cycle should correspond to the time point where the slope of the time course is close to 0. If more than one similar slope is available, choose the one which gives the titre that correlates best with the titre in TCID50. 6 . Choose the length of the virus replication cycle as the infection duration for the titration procedure.

Check that the time interval define in III is correct by comparing the virus titres obtained in the absence and in the presence of either an antibody (serum), or a virus entry inhibitors that block secondary infection added few hours after the virus inoculation (see text for details).

V Exclude false positive signal Inactivate your virus by UV treatment and check if you still get labelling by flow cytometry. If yes, virus titration by flow cytometry is not recommended.

Perform multiple titration of the same viral stocks using an already accepted technique i.e. TCID50/PFU and the novel method. 

Plaque formation and isolation of pure lines with poliomyelitis viruses

A simple method of estimating fifty percent endpoints

Improvements in methods for calculating virus titer estimates from TCID50 and plaque assays

Calculating HIV-1 infectious titre using a virtual TCID(50) method

A simple and rapid Hepatitis A Virus (HAV) titration assay based on antibiotic resistance of infected cells: evaluation of the HAV neutralization potency of human immune globulin preparations

Titration of human coronaviruses, HcoV-229E and HCoV-OC43, by an indirect immunoperoxidase assay

A rapid method for immunotitration of influenza viruses using flow cytometry

Production of glycoproteindeleted rabies viruses for monosynaptic tracing and high-level gene expression in neurons

Rapid titration of adenoviral infectivity by flow cytometry in batch culture of infected HEK293 cells

Rapid method for SV40 titration

Production, concentration and titration of pseudotyped HIV-1-based lentiviral vectors

Error assessment in recombinant baculovirus titration: evaluation of different methods

Rapid titration of viruses by flow cytometry

A flow cytometry-based immuno-titration assay for rapid and accurate titer determination of modified vaccinia Ankara virus vectors

SLAM (CDw150) is a cellular receptor for measles virus

Infection of chicken embryonic fibroblasts by measles virus: adaptation at the virus entry level

Inefficient measles virus budding in murine L.CD46 fibroblasts

End-point dilution and plaque assay methods for titration of cricket paralysis virus in cultured Drosophila cells

A European collaborative study to assess the proficiency of laboratory estimates of potency of live measles, mumps and rubella tri-valent vaccines

Identification of two amino acids in the hemagglutinin glycoprotein of measles virus (MV) that govern hemadsorption, HeLa cell fusion, and CD46 downregulation: phenotypic markers that differentiate vaccine and wild-type MV strains

CD46-mediated measles virus entry: a first key to host-range specificity

Efficient major histocompatibility complex class IIrestricted presentation of measles virus relies on hemagglutinin-mediated targeting to its cellular receptor human CD46 expressed by murine B cells

In vitro antiviral activity of dehydroepiandrosterone and its synthetic derivatives against vesicular stomatitis virus

Oncolysis of Hepatic Metastasis of Colorectal Cancer by Recombinant Vesicular Stomatitis Virus in Immune-Competent Mice

Primary human immunodeficiency virus type 2 (HIV-2) isolates, like HIV-1 isolates, frequently use CCR5 but show promiscuity in coreceptor usage

Characterization of the behavior of functional viral genomes during the early steps of human immunodeficiency virus type 1 infection

SIVSM/HIV-2 Vpx proteins promote retroviral escape from a proteasomedependent restriction pathway present in human dendritic cells

Identification of a methylated oligoribonucleotide as a potent inhibitor of HIV-1 reverse transcription complex

Potency estimation of measles, mumps and rubella trivalent vaccines with quantitative PCR infectivity assay

Estimation of the number of infectious measles viruses in live virus vaccines using quantitative real-time PCR

Quantitative Measurement of Varicella-Zoster Virus Infection by Semiautomated Flow Cytometry