1
Figure 7 (Supplementary Figure): Effect of the initial rate of death of infected hepatocytes on the predicted complexity of virus-cell junctions and hepatocyte turnover during virus clearance. Comp10 was used to calculate the effect of the initial rate of death of infected hepatocytes (percentage per first cycle) on predictions of (A) complexity of virus-cell junctions and (B) cumulative hepatocyte turnover after virus clearance. The initial rate of death of infected hepatocytes is shown as percent of infected hepatocytes in the first cycle assuming that the same number of hepatocytes was killed in all subsequent cycles (zero order). After each cycle, hepatocyte proliferation restores the liver to its original size. As shown, the predictions are relatively insensitive to differences in the initial rate of death up to about 35 percent.
Figure 8 (Supplementary Figure): Effect of delayed proliferation of hepatocytes on the predicted complexity of virus-cell junctions and hepatocyte turnover during virus clearance. Comp10 was used to calculate the effect of delayed hepatocyte proliferation on predictions of complexity (A) and cumulative hepatocyte turnover (B) after virus clearance. In this illustration, it was assumed that hepatocyte proliferation did not begin until a 50% decrease in liver size had occurred. Killing of infected hepatocytes was assumed to be zero order (initially 1% of infected hepatocytes) and proliferation of hepatocytes, once initiated, was assumed to be first order. Thus, when the liver size has dropped 50 percent, and hepatocyte proliferation initiates, 2% of surviving infected hepatocytes will be killed and the indicated percent will divide. The lowest rate of hepatocyte proliferation that was evaluated, 2.5% per cycle, just slightly exceeds the fraction of hepatocytes killed at this stage of the infection.
The amount of liver turnover for either Model 2 (cccDNA lost at mitosis) or Model 3 (cccDNA survives mitosis) stayed relatively constant, between 0.5-0.83 livers for Model 2, and 2.2 to 2.7 livers for Model 3, respectively. The only major effect of delayed hepatocyte proliferation was on the predicted complexity of virus-cell junctions, using Model 2. When the rate of hepatocyte killing was close to the rate of hepatocyte replacement, the complexity at recovery dropped from ~0.3 to 0.17; when the rate of hepatocyte replication was very high, the complexity at recovery increased from ~0.3 to 0.5.
We therefore asked how this idea of delayed hepatocyte proliferation would accommodate Model 2 (cf., Table 3), assuming a hepatocyte proliferation rate of 2.5 percent per cycle after the initial 50 percent decrease in hepatocyte number. Comp10 showed that the observed complexity of 0.73 for the ETV-treated group of woodchucks at recovery would correspond to an efficiency of detection of virus-cell junctions of 10%, with 0.83 liver turnovers, which would be sufficient for virus clearance and restoration of liver mass. To achieve the measured complexity of 0.41 for the untreated group of woodchucks, 12.7 liver turnovers would be needed, with the extra turnover attributed to a period of liver turnover without inhibition of virus replication prior to the clearance phase of the infection.
We also asked what would happen with Model 2 if the hepatocyte replication rate were 100 percent after the initial 50 percent decrease in hepatocyte number. In this case, Comp10 showed that the observed complexity of 0.73 for the ETV-treated group of woodchucks at recovery would correspond to an efficiency of detection of virus-cell junctions of 52%, with 0.51 liver turnovers, which would be sufficient for virus clearance and restoration of liver mass. To achieve the measured complexity of 0.41 for the untreated group of woodchucks, 1.6 liver turnovers would be needed, with the extra turnover attributed to a period of liver turnover without inhibition of virus replication occurring prior to the clearance phase of the infection.
In brief, if replication is delayed and subsequent liver regeneration is initially close to the rate of killing of infected hepatocytes, an extra-ordinary amount of liver turnover in the untreated woodchucks is needed to reconcile recovery by Model 2 (a 15-fold greater amount than in the ETV-treated group). When the rate of liver regeneration is very high, there is only a 3-fold difference in predicted liver turnover between the two groups.