PMC

Effect of duration of antigen stimulation on IFN-γ ELISpot test results. SFC counts (means of four replicates) in IFN-γ ELISpot following stimulation of PBMC samples from three CMV-seropositive healthy donors (d120, 32-year-old male; d254, 62-year-old female; d270, 22-year-old female) with T-activated® IE-1 or pp65 for 17, 19 and 21 h. Unstimulated PBMC (neg.) were used as a negative control. Differences between stimulation durations were tested using the non-parametric two-sided One-way ANOVA Kruskal-Wallis test (*P < 0.05). P-values < 0.05 were considered statistically significant

Comparison of IFN-γ ELISpot performance using different microtiter plates. ELISpot were performed using PBMC from three CMV-seropositive healthy donors (two preparations each, ELISpot in quadruplicate, ran twice) on various microtiter plate materials (MCE plate, PVDF plate, PVDF strips). Age (gender) of donors was 32 (male), 33 (male) and 51 (female). SFC values per 200,000 PBMC are shown as box plots for non-stimulated PBMC (neg.), and for T-activated® IE-1- and pp65-stimulated PBMC. Median values (horizontal black lines) are indicated in brackets. Differences between microtiter plate materials were tested using the non-parametric two-sided One-way ANOVA Kruskal-Wallis test (**P < 0.01; ***P < 0.001)

Determination of optimal concentrations of T-activated® pp65 and IE-1 for the stimulation of PBMC. PBMC were isolated from one CMV-seropositive healthy donor blood sample and stimulated for 19 h with increasing concentrations of T-activated® pp65 (31.6 fg/ml-31.6 μg/ml; black circles) or IE-1 (0.01-31.6 μg/ml; black squares). IFN-γ ELISpot results of four replicates are expressed as mean SFC/200,000 PBMC. Error bars represent standard deviations. The Y-axis scale was adjusted to pp65- (left) and IE-1- (right) specific values to optimize data resolution. Curve fitting was made using the four parameter logistic function of GraphPad Prism 5.04

Assay validation in immunocompetent donors. PBMC isolated from whole blood of 45 healthy donors (Table ) were assayed using the optimized IFN-γ ELISpot. A positivity cut-off of 10 SFC/200,000 PBMC (grey horizontal dashed line) was defined (see text and Additional File ). Considering a test result as positive when the geometric mean for at least one of the IE-1 or pp65 stimulated approach is ≥ 10 SFC/200,000 PBMC and when the ratio of geometric means of stimulated to unstimulated conditions is ≥ 2.5, positive agreement (sensitivity) and negative agreement (specificity) of the optimized IFN-γ ELISpot test results with CMV serology within this collective of healthy donors was 97% and 85% respectively

titration of the ifn-γ capture antibody. PBMC (four replicates each) from five CMV-seropositive healthy donors (median age and range of 31 (22–49) years; 1 male and 4 female) were seeded on PVDF microtiter strips coated with increasing concentrations (2.5 to 7.5 μg/ml) of IFN-γ capture antibody. Cells were left unstimulated (neg.) (a, b) or were stimulated with T-activated® IE-1 (a) or pp65 (b, c) CMV antigens, as before. SFC mean values per 200,000 PBMC are shown as box plots for the collective of 5 donors (median values indicated in brackets) (a, b) and as histograms (c) for pp65-stimulated PBMC of individuals donors (d032, d120, d172, d202, d241). Differences between coating conditions were tested using the non-parametric two-sided One-way ANOVA Kruskal-Wallis test (*P < 0.05; **P < 0.01)

IFN-γ ELISpot assay linearity. a Working range of PBMC per ELISpot assay. Increasing number of PBMC from one CMV-seropositive healthy donor were seeded per well and ELISpot assays were performed as described, following stimulation with T-activated® IE-1 or pp65. Mean SFC values and standard deviation obtained for 60,000-200,000 PBMC per well are depicted. Scale of the Y-axes was adjusted for pp65 (left) and IE-1 (right) for a better data resolution. b Linearity between the number of CMV-reactive PBMC and enumerated SFC. The indicated numbers of PBMC from one CMV-seropositive healthy donor were mixed with PBMC from one CMV-seronegative donor (up to 200,000 total PBMC) and stimulated with T-activated® pp65 antigen according to the optimized protocol. Mean SFC values and standard deviation of quadruplicate measurements are shown for two donor pairs (d034 + d219, d204 + d067). In both panels, regression lines and corresponding coefficient of determination R² were generated using the regression line tool of GraphPad Prism 5.04

Determination of assay sensitivity and specificity. PBMC from 10 each CMV-seropositive and CMV-negative healthy blood donors were left unstimulated (neg.) or were stimulated with T-activated® pp65 or IE-1 and IFN-γ ELISpot assays were conducted as before. Mean SFC values per 200,000 PBMC of 4 replicates are shown as box plots. Median values (horizontal black lines) are indicated in brackets. Y-axis scales were adjusted in each graph for better resolution of SFC counts. Median age and range of CMV-seronegative and CMV-seropositive subjects was 28 (24–53) and 31 (23–56) years. Gender distribution in CMV-seronegative (30% male and 70% female) and CMV-seropositive (25% male and 75% female) groups was comparable. Differences between unstimulated and stimulated conditions were tested using the non-parametric two-sided Mann–Whitney U (MWU) test. P-values < 0.05 were considered statistically significant

Evaluation of different cell culture media on the performance of IFN-γ ELISpot. a Boxplot diagram showing ELISpot results upon stimulation of PBMC of 10 CMV-seropositive healthy blood donors (median age and range of 31 (26–54) years; 40% male and 60% female) with T-activated® pp65 in various cell culture media. Median values (horizontal black lines) are indicated in brackets. Differences between medium conditions among stimulated conditions were tested using the non-parametric two-sided One-way ANOVA Kruskal-Wallis test (p = 0.613). b ELISpot results of the non-stimulated cells incubated in AIM-V or in UltraCulture (UC) serum-free media are shown separately with an expanded Y-axis scale. Differences between both conditions were tested using the non-parametric two-sided MWU test (p = 0.074). P-values < 0.05 were considered statistically significant. Serum-containing media were composed of RPMI 1640 supplemented with 5% serum (R5): FCS, NTA, NTS or human AB (hAB)

T-activated® pp65 and IE-1 CMV antigens stimulate a broad range of CMV-reactive effector cells. Comparative analysis of IFN-γ secreting cells by ELISpot (a) and flow cytometry (b) following stimulation of PBMC with T-activated® pp65 and IE-1 antigens. PBMC (4 replicates each) from six CMV-seropositive healthy donors (median age and range of 39 (22–55) years; 2 male and 4 female) were stimulated with T-activated® pp65 and IE-1 antigens according to the optimized IFN-γ ELISpot (a) and to the protocol of intracellular and surface marker staining and flow cytometry described in the Methods section (b). Bar graphs in (a) depict IFN-γ-dependent SFC per 200,000 PBMC, as before. Bar graphs in (b) represent the number of IFN-γ-expressing CD3⁺CD4⁺ (Th), CD3⁺CD8⁺ (CTL), CD3⁻CD56⁺ (NK) and CD3⁺CD56⁺ (NKT-like) cells per 200,000 lymphocytes. Note that the Y-axis scales in (b) were adjusted for a better resolution of the respective data. Individual age and gender of donors were as follows: d120, 37-year-old male; d172, 55-year-old female; d290, 46-year-old female; d300, 41-year-old female; d343, 24-year-old female; d361, 22-year-old male