Validating the Guava ViaCount® Assay, an Automated Cell Counting/Viability Method for Use in ELISpot Assays

Abstract

Many of the standard methods currently used for cell counts and viability assessments of cell suspensions, e.g., peripheral blood mononuclear cells (PBMC), do not provide very accurate or reliable measurements. The Guava PCA™ System with the ViaCount assay pro-vides enhanced sensitivity and better precision than optical tools such as the hemocytometer for the cell counting and viability determination. A major benefit of the Guava PCA is that the cell counting process can be automated, thereby reducing the time and effort that must be spent for the analysis. In addition, this instrument eliminates a significant degree of operator judgment and simplifies the validation of the method. However, adopting a new method (instrument and assay) into the GLP environment requires quality testing and assay optimization to develop standard operating procedures which must then be validated. In this application note, we describe the use of a human IFN-γ ELISpot to assess the precision of fluorescent cell counting using capillary microcytometry as part of the validation process for the Guava ViaCount assay.

These validation experiments focused on comparing cell counts performed with trypan blue exclusion (hemocytometer) and the Guava ViaCount Assay. The precision of the counts was then evaluated with three IFN-γ ELISpot Assays on PBMC preparations from five different donors. The correlation between cell counts and spot forming cells (SFC) were used to judge the precision and reproducibility of the hemocytometer and Guava ViaCount Assay.

Guava ViaCount results showed a good correlation to the manual counts, with a correlation coefficient (r) of 0.88. Guava ViaCount assay cell count results had significantly lower CV’s than the manual trypan blue method. When the ELISpot assay was used, we observed that the variability seen in the PHA response (SFC counts) between each pair of PBMC samples (counted by hemocytometer and ViaCount assay) run on the same plate was directly related to the difference in each pair of cell counts. The inter-assay precision seen in the PHA response among all three experiments was better with less variability when the samples were diluted based on ViaCount Assay results.

Our data demonstrated that the results of the Guava ViaCount assay are comparable to those obtained from the manual procedure in which a hemocytometer is used with the trypan blue dye. The Guava ViaCount assay, however, is readily automated and provides more consistent results.

Introduction

The ELISpot (Enzyme-Linked Immunospot) technique has been shown to be an effective measure of cell-mediated immune responses.1 When the technique was first developed, it was mainly used to detect antibody-secreting cells.2,3 In recent years, the technique has been adapted for the detection of cytokines secreted by T cells.4,5 An important application of the ELISpot technique is monitoring the effectiveness of vaccines used in clinical trials.6,7 When a patient is vaccinated, antigen-specific T cells are programmed to recognize and respond to the antigen (virus or other foreign agent). The ELISpot assay is therefore a useful method in monitoring the patient’s immune response to the immunizing antigen and an important marker of cellular mediated immunity. The number of T cells stimulated to produce cytokine detected by the ELISpot assay indicates the patient’s response to vaccination.

The steps included in the ELISpot technique are as follows:

a) The well membrane is coated with the appropriate antibody to a specific cytokine.

b) The peripheral blood mononuclear cells (PMBC) are added to the well along with a stimulating antigen and incubated on the membrane.

c) Secreted cytokines are captured by the membrane bound antibody in the vicinity of the producing cell. Cells are removed by washing.

d) Biotinylated antibodies to the cytokine are added to form an antibody-cytokine-antibody “sandwich”.

e) The avidin-enzyme conjugate is added and binds to the biotinylated antibody. A colored substrate which reacts with the enzyme conjugate and precipitates at the location of the cytokine producing cell is added to the membrane.

f) The spots on the membrane develop and the plate is dried. Each spot is counted.

A critical step in the assay is monitoring PBMC sample integrity and accurately counting the cells prior to stimulation in the wells on the ELISpot plate. Accurate and precise cell counts will affect ELISpot outcomes, particularly when comparing multiple timepoints, donors, operators and laboratories. Many laboratories use a hemocytometer and a trypan blue exclusion test as a means of counting viable cells. While this procedure is relatively simple, it suffers from a number of deficiencies due to the manual aspects of the procedure. Since it is subjective and difficult to differentiate apoptotic cells from live and dead cells, trypan blue exclusion can incorrectly estimate viabilities.8

Sample integrity can also impact the ELISpot results. Contaminating red blood cells (RBC) and platelets not removed from the sample can adversely affect T cell function;9 freeze/thaw techniques could increase the amount of debris in the sample; and apoptotic and/or dead cells will not respond in an ELISpot assay. In addition, the manual procedure does not lend itself to monitoring the samples for RBC, platelets or debris contamination. There is a high demand coming from laboratories to find an automated cell counting system providing viable cell counts, with high speed, accuracy and reproducibility. Flow cytometry with cell fluorescence labeling, i.e., Guava PCA System, is a powerful alternative to optical methods for viable cell counting. The Guava ViaCount assay distinguishes between viable and non-viable cells based on the differential permeability of DNA-binding dyes in the ViaCount Reagent. One dye is membrane permeable and stains all nucleated cells (PM2). The other dye only penetrates cells with compromised membrane integrity (PM1). Events are counted if they emit nucleated cell fluorescent signal and the forward light scatter (FSC) intensity is appropriate for a particle the size of a cell. In this system, debris events with low FSC signal are not counted. This application note describes the use of fluorescence based capillary microcytometry to count cells before stimulation in the IFN-γ ELISpot assay.  We present results from three validation experiments comparing cell counts performed with the hemocytometer and the Guava ViaCount assay. The interassay variation among the IFN-γ ELISpot runs and the correlation between cell counts and SFC were used to judge the precision and reproducibility of the Guava ViaCount Assay.

Materials and Methods

Cell sample preparation

The PBMC samples used in the validation experiments were isolated from leukocyte cell packs collected from healthy donors and purchased from BRT Laboratories, Inc. (Baltimore, Maryland). Each leukocyte cell pack was collected, processed using Ficoll-Paque gradient, and cryogenically preserved the same day with 22% fetal bovine serum (FBS) and 7.5% dimethyl sulfoxide (DMSO). These frozen PBMC were then stored in the vapor phase of liquid nitrogen (−150 °C) until testing. Each assay evaluated cells from five different donors. For each donor a vial was thawed and the contents suspended in chilled RPMI 1640 (Mediatech, Cat. 10-041-CM) supplemented with 10% FBS and 50U/mL of DNAse I (Sigma, Cat. D7291). The cells were pelleted by centrifugation then suspended in chilled RPMI 1640 with 1% fetal bovine serum and 50 U/mL of DNAse I. Following another round of centrifugation, the cell pellets were suspended in RPMI 1640 with 10% fetal bovine serum. Each cell suspension was allowed to recover overnight in a humidified 5% CO2 incubator at 37 °C. The cells were then washed once and suspended in RPMI 1640 with 10% fetal bovine serum for use in the ELISpot assay.

Manual cell counting and viability

A Neubauer-type Hemocytometer counting chamber (Hausser Scientific, Cat. Num. 1475) and a contrast-phase microscope (Olympus model CX41) were used for manual cell counts. A 20 µL aliquot of each cell suspension was diluted by mixing with 80µL of ACK Lysing Buffer (Quality Biological Cat. #118-156-061) containing 0.1% trypan blue solution (Sigma Cat. T8154). Samples treated with trypan blue (viable cells exclude the dye) were loaded onto each side of the counting chamber and counted using a contrast phase microscope. A viable cell count for each donor cell suspension was determined by counting the number of live cells in the 16 smaller squares in each of the four large squares on the hemocytometer. The total viable cell count was determined by multiplying the number of viable cells counted and the appropriate dilution factors. The number of live (clear) and dead (blue) cells was recorded and the percentage of live cells calculated.

Automated cell counting with the Guava ViaCount assay and daily Instrument Performance Check

The same cell suspensions used to perform manual cell counts were used to prepare stained automated cell counts using the Guava ViaCount reagent (Guava Cat. #4000-0040). Following the recommended procedure in the Guava ViaCount reagent package insert, 20 µL of each cell suspension was mixed with 380 µL of Guava ViaCount reagent in a 1.5 mL microcentrifuge tube. This 20-fold dilution was then incubated between 5 and 30 minutes at room temperature prior to analysis with the Guava PCA system. Before the ViaCount assay was run, the instrument went through a daily performance check using the Guava Check application and the Guava Check reagents according to the product insert and the Guava PCA User’s Guide. After passing the performance check, each ViaCount Assay sample was analyzed with the Guava PCA instrument. Relative information such as Viable cells per milliliter, Viability %, and Total Viable Cells in the original sample were obtained from the Data Analysis Screen.

Cell sample stimulation and assay plate design for the IFN-γ ELISpot Assay

Following thawing, washing and overnight incubation as described above, each donor cell suspension was split into two equal aliquots. Each pair of aliquots was diluted to a concentration of 1x106 cells/ mL. One aliquot was diluted according to calculations from the manual trypan blue cell count. The other was diluted based on the calculations from the automated ViaCount Assay cell count. Each ELISpot assay experiment was performed using a separate set of thawed cells from the five specified donors. Each plate was divided evenly between wells receiving cell samples whose concentrations were adjusted based on the manual cell counts or on the Guava ViaCount Assay results. The cell suspensions were adjusted to a concentration of 1x106 cells/mL and 50 µL (50,000 cells) of each cell suspension was added per well. Each cell sample was stimulated with Phytohemagglutinin PHA-P (Sigma Cat. L8754) by addition of 50 µL of PHA-P (8 µg/mL) was added per well. The plates were placed in a humidified 5% CO2 incubator at 37 °C for 18 to 20 hours.

Plate development and IFN-γ detection

The next day, the assay plates were developed using the detection reagents contained in the BBI Biotech Human IFN-γ ELISpot Assay Kit (BBI Biotech Cat. No. 68000). The kit contains the following reagents: 20X Wash Solution, 5X Diluent Buffer, 100X Detection Antibody, 100X Conjugate, Substrate Reagent 1, Substrate Reagent 2, Substrate Reagent 3, and Substrate Reagent 4. Each plate was washed six times with wash solution, and then each plate received 50 µL/well of Detection Antibody Solution. The plates were incubated for two hours in a humidified 5% CO2 incubator at 37 °C and washed six times with wash solution. Next, 100 µL/well of Conjugate solution was added to the plates. The plates were covered and incubated for one hour at room temperature and washed again six times with wash solution. Fresh substrate solution was prepared by mixing 75 µL of Substrate Reagent 1, and 50 µL each of Substrate Reagent 2, Substrate Reagent 3, and Substrate Reagent 4 with 20 mL of autoclaved water. Each plate received 100 µL/well of the substrate solution. The plates were allowed to incubate at room temperature in the dark for five minutes and then immediately rinsed with cool tap water for five minutes and dried overnight in the dark before analysis.

ELISpot Assay plate analysis and acceptance criteria

The following day, the wells from the completed assay plates were scanned and the spots were counted with an ImmunoSpot® Analyzer (Cellular Technologies Ltd.). Each plate was analyzed with ImmunoSpot Software version 2.08 with the following settings:

Sensitivity = 150

Background Balance = 80

Diffuse Spot Process = off

Minimum Spot Size = 0.0001 mm2

Maximum Spot Size = 2.7759 mm2

Spot Separation = 3

Overdeveloped Area = Active

Overdeveloped Area Processing = Estimate

Oversized Spots = Estimate.

The BBI Biotech Human IFN-γ Elispot acceptance criterion requires that 90% of the used wells from each assay plate must have a spot forming cell (SFC) count per 50,000 cells equal to or greater than 200 in the Positive [PHA-P stimulated] control wells and the mean number of SFC per 50,000 cells for the Negative [unstimulated] wells must be less than 5 on each assay plate. A total of three ELISpot assays plates were run for this validation. Donor B from Plate 1 was eliminated due to high back-ground. The remaining four donors on Plate 1 and all five donors on Plate 2 and 3 met the acceptance criteria.

Results

Viable Cell Count Studies

To determine if there were differences in the cell counts provided by the two methods, each thawed PBMC sample was washed and then counted with both the trypan blue exclusion test and Guava ViaCount Assay. A summary of the viable cell counting results are pre-sented in Table 1. The Guava ViaCount yielded higher cell counts than trypan blue exclusion (p< 0.001, paired T-test, n=15) and a lower coefficient of variability (p< 0.05, T-test, n=5). The Guava ViaCount results however did show a strong correlation to the manual counts with a correlation coefficient (r) of 0.88 (Figure 1). To deter-mine whether the difference in the two cell counting methods was dependent on cell number, the difference between the two counts was analyzed in relationship with the mean cell count. The difference appeared equal over the range of cell counts (Figure 2). This figure also confirmed that cell counts from the Guava ViaCount Assay tended to be greater than trypan blue counts.

Table 1. Summary of Total Viable Cell Count Data

Donor ID Cell Counting Method Plate 1 Plate 2 Plate 3 Plate 4 Plate 5
Donor A Trypan Blue 1.20E+07 1.30E+07 1.43E+07 1.31E+07 8.80
ViaCount 1.51E+07 1.38E+07 1.38E+07 1.42E+07 5.27
Donor B Trypan Blue 4.57E+06 5.20E+06 6.50E+06 5.42E+06 18.15
ViaCount 8.14E+06 9.26E+06 8.22E+06 8.54E+06 7.32
Donor C Trypan Blue 9.60E+06 9.00E+06 5.50E+06 8.03E+06 27.56
  ViaCount 1.20E+07 8.14E+06 8.07E+06 9.40E+06 23.92
Donor D Trypan Blue 6.60E+06 6.60E+06 4.20E+06 5.80E+06 23.89
ViaCount 7.34E+06 6.65E+06 6.96E+06 6.98E+06 4.95
Donor E Trypan Blue 3.95E+06 6.80E+06 4.10E+06 4.95E+06 32.40
ViaCount 6.89E+06 6.21E+06 5.90E+06 6.33E+06 8.00

Figure 1: Correlation of Trypan Blue/Manual Count vs. Guava ViaCount Assay.

Figure 1: Correlation of Trypan Blue/Manual Count vs. Guava ViaCount Assay.

Relationship of the Difference in Viable Cell Counts to the Mean Viable Cell Count for both Methods of Counting.

Figure 2: Relationship of the Difference in Viable Cell Counts to the Mean Viable Cell Count for both Methods of Counting. The total viable cell count obtained using the trypan blue staining was subtracted from the total viable cell count obtained from the Guava ViaCount for the same donor cells. The difference was graphed vs. the mean value for the two counts. Values are as total viable cell counts. The greater the distance from zero, the greater the difference between the two cell counts. All the dots above zero represent Guava Counts that were greater than trypan blue counts.

In summary, the value from the Guava ViaCount was correlated with the value from the trypan blue method employing a hemocytometer, however, the Guava ViaCount yielded higher viable cell counts. In addition, the data collected from the Guava ViaCount had less variability than the data from the hemocytometer.

Variability seen in donor responses to stimulation

To evaluate whether differences in the cell counts effected the outcome of subsequent bioassays, ELISpots were performed independently on both cell counting methods. Before each pair of samples was stimulated and tested via the ELISpot assay, they were diluted to a concentration of 1 x 106 cells/mL and 50,000 cells added per well. There was more variability observed in the SFC counts among the 3 assays when hemocytometer counts were used to determine cell concentration than Guava ViaCount (Figure 3). This resulted in lower interassay coefficient of variability and better precision when Guava ViaCount numbers were used for setting up the ELISpot (Table 2).

Figure 3: PHA Response of Five Donors in Three ELISpot Assay Plates. PBMC were counted using a hemocytometer with trypan blue staining and the Guava ViaCount assay.

Figure 3: PHA Response of Five Donors in Three ELISpot Assay Plates. PBMC were counted using a hemocytometer with trypan blue staining and the Guava ViaCount assay. Three independent ELISpot assays were performed with four replicate wells per donor per condition. PBMC’s suspensions diluted based on trypan blue viability count were plated on the left side of each plate. PBMC’s suspensions diluted based on the Guava ViaCount viability count were plated on the right side of the same plate. Assays were performed 1 to 3 days apart.

Table 2. Analysis of ELISpot Assay Results (SFC per 50,000 cells)

Hemocytometer Wells
ViaCount Wells
Donor ID Mean Std. Dev. %CV Mean Std. Dev. %CV
Donor A 311.0 61.7 19.8 298.9 64.0 21.4
Donor B 378.4 27.6 7.3 307.3 54.4 17.7
Donor C 362.7 130.3 35.9 315.2 57.7 18.3
Donor D 272.8 94.8 34.7 223.5 23.3 10.4
Donor E 538.2 106.7 19.8 486.1 22.9 4.7

Discussion and Conclusions

The Guava ViaCount technique, which is readily automated, yields results that are comparable to those obtained from the manual procedure using a hemocytometer and the trypan blue exclusion test. The data clearly indicate that the Guava ViaCount assay can provide reliable cell counting data with a greater degree of precision than the manual hemocytometer, thereby improving the utility of the results from the ELISpot assay. Also, the ViaCount Assay has additional benefits by allowing an operator to monitor sample integrity for apoptotic cells or contamination from RBC, platelets or debris. In addition to providing cell viability numbers and percentages in an easy to read format, the ViaCount Assay has a number of additional benefits. As an example, it allows the operator to easily monitor samples for possible platelet contamination, which can affect Cytokine measurements or cell counts (see Figure 4).

Figure 4: The Guava ViaCount Analysis Display presents a clear, concise overview of the assay, indicating the analytical results. In addition, the live cells and dead cells can be identified, as shown in (A) with platelets gated out, and (B) with platelets visible.

Figure 4: The Guava ViaCount Analysis Display presents a clear, concise overview of the assay, indicating the analytical results. In addition, the live cells and dead cells can be identified, as shown in (A) with platelets gated out, and (B) with platelets visible.

In summary, the Guava ViaCount results demonstrated a high correlation with the manual hemocytometer counts, with counts routinely greater than those from a hemocytometer. When the ELISpot assay was per-formed the variability seen in the PHA response was less when samples were diluted based on ViaCount results. The results of this validation experiment support the use of the Guava PCA system to provide cell viability counts comparable to the manual trypan blue exclusion method with improved precision and less variability.

Materials

     

References

  1. Czerkinsky, C.C.; Nilsson, L.A.; Nygren, H.; Ouchterlony, O.; Tarkowski, A. A solid-phase enzyme-linked immunospot (ELISPOT) assay for the enumeration of specific antibody secreting cells. J Immunol. Methods 1983; 65:109–121.
  2. Sedgwick, J.D.; Holt, P.G. A solid-phase enzyme-linked immunoenzymatic technique for the enumeration of specific antibody-secreting cells, J Immmunol. Methods 1983; 57:301–309.
  3. Czerkinsky, C.C.; Tarkowski, A.; Nilsson, L.A.; Ouchterlony, O.; Nygren, H.; Gretzer, C. Reverse enzyme-linked immunospot assay (RELISPOT) for the detection of cells secreting immunoreactive substances, J Immunol. Methods 1984; 72:489–496.
  4. Carvalho, L.H.; Hafalla, J. C. R.; Zavala, F. ELISPOT assay to measure antigen-specific murine CD8+ T cell responses. J Immunol. Methods 2001; 252:207–218.
  5. Rininsland, F. H.;Helms, T.; AsaaD, R. J.; Boehm, B. O.; Tary-Lehmann, M. Granzyme B ELISpot assay for ex vivo measurement of T cell immunity. J Immunol. Methods 2000; 240: 143–155.
  6. Mwau, M.; McMichael, A. J.; Hanke, T. Design and Validation of an Enzyme-Linked Immunospot Assay for use in Clinical Trials of a Candidate HIV Vaccines, AIDS Res. Hum. Retro. 2002; 18; 9: 611–618.
  7. Cox, J.; Ferrari, G.; Kalams, S.A.; Lopaczynski, W.; Oden, N.; D'Souza, M.P. ELISPOT Collaborative Group Results of an ELISPOT Proficiency Panel Conducted in 11 Laboratories Participating in International Human Immunodeficiency Virus Type 1 Vaccine Trials. AIDS Res. Hum. Retro. 2005; 21; 1: 68–81 (in press).
  8. Mascotti, K.; McCullough, J.; Burger, S.R. HPC viability measurement: trypan blue versus acridine orange and propidium iodide. Transfusion 2000; 40: 693–696.
  9. Song, S.; Goodwin, J.; Zhang, J.; Babbitt, B.; Lathey, J.L. Effect of Contaminating Red Blood Cells on OKT3-Mediated Polydonal Activation of Peripheral Blood Mononuclear Cells. Clin. Diagn. Lab. Immunol., 2002; 9:708–712.