Fig. 1. Measurements of HSV-1 DNA yield correlate with viral titers. (A) HSV-1 titers were determined by plaque assay in Vero cells 24 h after inoculation with 0.002 to 4.6 PFU per cell (n=2 per MOI). (B) Dotblots of DNA samples harvested from uninfected Vero cells (MOI=0) and HSV-1 infected Vero cells 24 h after inoculation (n=2 per MOI). Estimates of HSV-1 DNA yield were calculated based on the amount of HSV-specific oligonucleotide that hybridized to each DNA sample. (C) The log (viral DNA yield) is plotted as a function of MOI, and is expressed in terms of log (fold-increase above the lower limit of detection); thus, the value ‘0’ on the y-axis indicates the lower limit of detection of the assay.

Fig. 2. Effect of human IFN-β on HSV-1 replication. (A) and (B) Dotblot of DNA samples harvested (A) 30 or (B) 48 h after inoculation with HSV-1 (MOI=0.1). Vero cells were treated with half-log dilutions of human IFN-β ranging in concentration from 0.03 to 10,000 U/ml (n=2 per concentration). HSV-1 DNA yields were measured based on the efficiency with which an HSV-specific oligonucleotide probe hybridized to each DNA sample. The log (viral DNA yield) measured (C) 30 or (D) 48 h after inoculation is plotted as a function of the log [IFN-β]. Open diamonds and error bars indicate the mean±SD of log (viral DNA yield), which is expressed in terms of log (fold-increase above the lower limit of detection); thus, the value ‘0’ on the y-axis indicates the lower limit of detection.

Fig. 3. Mathematical description of inhibition of HSV-1 replication by IFN-β. (A) The relationship between V (log [viral DNA yield]) and I (log [IFN]) is dictated by 1. ΔI, one-half the width of the difference between I_min and I_max, 2. ΔV, the height of the difference between V_max and V_min, and 3. the midpoint of the dose–response curve, I₅₀, V₅₀ (closed circle). The locations of I_min and I_max are denoted by vertical dotted lines and the open circles on the dose–response curve. The upper and lower asymptotes of the hyperbolic tangent function, V_max and V_min respectively, are denoted by horizontal dashed lines. (B) and (C) Relationship of log (viral DNA yield) to log [IFN-β] when measured (B) 30 or (C) 48 h after inoculation. Open diamonds and error bars indicate the mean±SD of measured values of log (viral DNA yield), which are expressed as log (HSV-1 DNA). The dashed line indicates the simulated values that were derived by fitting the V(β) equation to visual estimates of V_min, V_max, I_min, and I_max. The solid line indicates the simulated values derived by fitting the equation for (B) V(β)^30 h and (C) V(β)^48 h to their respective data sets by the method of least squares. The maximum reductions in log (viral DNA yield) achieved by IFN-β are denoted by the vertical arrows labeled and .

Fig. 4. Effect of human IFN-γ on HSV-1 replication. (A) and (B) Dotblot of DNA samples harvested (A) 30 or (B) 48 h after inoculation (MOI=0.1). Vero cells were treated with half-log dilutions of IFN-γ ranging in concentration from 0.03 to 10,000 U/ml (n=2 per concentration). (C) and (D) Relationship of log (viral DNA yield) to log [IFN-γ] when measured (C) 30 or (D) 48 h after inoculation. Open triangles and error bars indicate the mean±SD of measurements of log (viral DNA yield), which are expressed in terms of log (fold-increase above the lower limit of detection); thus, the value ‘0’ on the y-axis indicates the lower limit of detection. The locations of I_min and I_max are denoted by vertical dotted lines, and V_max and V_min are denoted by horizontal dashed lines. The closed circle in the middle of the square indicates the location of I₅₀, V₅₀. The solid line indicates the simulated values derived by fitting the equation for (C) V(γ)^30 h or (D) V(γ)^48 h to their respective data sets by the method of least squares. The maximum reductions in log (viral DNA yield) achieved by IFN-γ are denoted by the vertical arrows labeled and .

Fig. 5. Additive composite curve analysis of the combined effects of IFN-β and IFN-γ. (A) and (B) Dotblot of DNA samples harvested (A) 30 or (B) 48 h after inoculation (MOI=0.1). Vero cells were treated with half-log dilutions of human IFN-β, human IFN-γ, or 1:1 combinations of IFN-β and IFN-γ that ranged in concentration from 0.03 to 10,000 U/ml (n=2 per concentration). (C) and (D) Relationship of log (viral DNA yield) to log [IFN-β] (open diamonds), log [IFN-γ] (open triangles), or log [IFN-β+IFN-γ] (open circles) when measured (C) 30 or (D) 48 h after inoculation. Open symbols and error bars indicate the mean±SD of measurements of log (viral DNA yield), which are expressed in terms of log (fold-increase above the lower limit of detection); thus, the value ‘0’ on the y-axis indicates the lower limit of detection. For the V(β:γ) functions, the locations of I_min and I_max are denoted by vertical dotted lines, V_max and V_min are denoted by horizontal dashed lines, and the closed circle denotes the location of I₅₀, V₅₀. The line of X's indicates the reduction in log (viral DNA yield) predicted by the null (dose-additive) hypothesis. The solid black line indicates simulated values derived by fitting the equations (C) V(β:γ)^30 h or (D) V(β:γ)^48 h to the observed data. The maximum observed reductions in HSV-1 DNA yield achieved by 1:1 combinations of IFN-β+IFN-γ are denoted by the vertical arrows labeled and .

Fig. 6. Response surface analysis of the combined effects of IFN-β and IFN-γ. (A) and (D) Dotblot of DNA samples harvested (A) 30 h or (D) 48 h after inoculation (MOI=0.1). Vero cells were treated with a matrix of 1/3rd log dilutions of IFN-β and IFN-γ that ranged in concentration from 0 to 1000 U/ml. (B) and (E) Relationship of log (HSV-1 DNA) to the combined concentration of IFN-β and IFN-γ in cultures treated with increasing doses of IFN-β and constant doses of IFN-γ of either 0, 21, or 100 U/ml, as determined (B) 30 h or (E) 48 h after inoculation. The equations V(β) and V(β:γ) provide reference points for previously observed effects of IFN-β versus 1:1 combinations of IFN-β and IFN-γ, and are based on the terms presented in Table 1. (C) and (F) The three-dimensional response surface of fold-reduction in HSV-1 DNA yield, as determined (C) 30 h or (F) 48 h after inoculation. Fold-reduction is defined as HSV-1 DNA yield in ‘vehicle-treated cells’/‘IFN-treated cells.’ Isoboles (contours) on the response surface occur in 20-fold increments.

Fig. 7. Mathematical models of the interaction between IFN-β and IFN-γ. (A) and (B) Predictions of the dose-additive model (Eq. (5)). Three-dimensional response surfaces of fold-reduction in viral DNA yield predicted to occur (A) 30 h and (B) 48 h after inoculation if IFN-β and IFN-γ interact in a dose-additive manner. Isoboles on the response surface occur in 20-fold increments. (C) and (D) Predictions of the multiplicative interaction model (Eq. (6)). Three-dimensional response surfaces of fold-reduction in viral DNA yield that are predicted to occur (C) 30 h and (D) 48 h after inoculation if IFN-β and IFN-γ interact in a multiplicative manner. (E) and (F) Observed measurements of log (viral DNA yield) obtained (E) 30 h and (F) 48 h p.i. are plotted on the x-axis relative to the values predicted by the dose-additive and multiplicative interaction models (y-axis). The line of unity in each graph denotes the series of points at which there is a perfect correlation between predicted and observed values.

Individual versus combined effects of IFN-β and IFN-γ on HSV-1 replication