Fig 1.
The structure of PCNA and the docking of small molecule inhibitors.
A) The structure of the PCNA homotrimer encircling double stranded DNA (PDB file 6GIS) [79]. Black boxes highlight the areas where PCNA-I1 and T2AA bind. B) PCNA-I1 is predicted to bind to the interface between two PCNA monomers by molecular modeling [40]. Specific residues that are predicted to interact with PCNA-I1 are labeled in light blue. D86 of one PCNA monomer (pink) is predicted to bind to the PCNA-I1 inhibitor via a N-O hydrogen bond. Residue K110 of the same monomer (pink) is predicted to form a nonpolar bond via the aromatic rings of PCNA-I1. R146 on an adjacent PCNA monomer (yellow) is modeled to bind to PCNA-I1 via an O-N hydrogen bond. C) The chemical structure of PCNA-I1. D) Structure of T2AA bound to PCNA. It was found that two molecules of T2AA bind to a PCNA monomer (blue) (PDB file 3WGW) [45]. T2AA binds within the interdomain connecting loop where proteins containing the PIP-box motif interact. The second T2AA molecule binds to the PCNA monomer adjacent to K164 (residue labeled in orange). E) Chemical structure of T2AA. Images of PCNA created with PyMOL.
Fig 2.
Effect of PCNA inhibitors on cell viability, DNA replication, and PCNA protein levels.
A-B) The Cell Titer Glo Luminescent Cell Viability assay was carried out to measure the cytotoxic effects of PCNA-I1 (A) and T2AA (B) on MRC-5 and Vero cells. Cells were incubated in the presence of indicated concentrations of inhibitor for 24 hours before conducting the assay. Data represent the mean of biological triplicate experiments with standard deviation. Each biological replicate was determined as the mean of 3 technical replicates. C) Immunofluorescence images of MRC-5 cells that were plated at a low density and uninhibited or treated with 2.5 μM PCNA-I1 or 12.5 μM T2AA for 6 hours. After 4 hours, EdC was incorporated into replicating cellular DNA for two hours. Following fixation, nuclei were stained with Dapi, EdC-labeled DNA was tagged with Alexa Fluor 488, and PCNA was stained by immunofluorescence. Scale bars, 50 μm. D) The percentages of replicating cells after PCNA-I1 or T2AA inhibition were calculated from 10 images captured as in (C). The number of nuclei with EdC incorporation divided by total nuclei counted is indicated above each bar graph. E) The CellTiter-Glo Luminescent Cell Viability assay was carried out as in (B) in the presence of cisplatin. Cells were incubated in the presence of the indicated concentration of cisplatin and/or 12.5 μM T2AA for 24 hours before conducting the assay. Data represent the mean of biological triplicate experiments with standard deviation. Each biological replicate is an average of three technical replicates. One-way ANOVA with the Dunnett’s multiple comparisons test was performed to compare cell viability in the presence of 100, 200, or 300 nM cisplatin to the condition with no cisplatin added (0 nM). In the absence of T2AA, cisplatin had little effect on cell viability (ns = nonsignificant). T2AA treatment sensitized cells to cisplatin treatment (* p < 0.1, 90% confidence interval; ** p < 0.05, 95% confidence interval; *** p < 0.01, 99% confidence interval; **** p < 0.001, 99.9% confidence interval). F) Western blots of whole-cell lysates collected from PCNA-I1 or T2AA treated MRC-5 cells, or uninhibited (UI). Total protein was collected at 2, 4, 6, 8, and 24 hours post inhibitor addition. Blots were probed with α-PCNA or α-GAPDH antibody.
Fig 3.
PCNA inhibitors inhibit HSV-1 infection.
A-B) Effects of inhibitors on low multiplicity HSV-1 infection. Cells were supplemented with either PCNA-I1 (A) or T2AA (B), or as a control, uninhibited. Inhibitors were added one hour before and throughout infection. One million MRC-5 cells were infected at an MOI 0.1 PFU/cell in the presence or absence of PCNA-I1 or T2AA. Virus was collected 24 hours later. Viral yield was determined via plaque assay in Vero cells. All values represent the means of biological duplicate experiments with standard deviations. Data points without observable error bars represent highly reproducible data. C) Effects of inhibitors on high multiplicity infection. Experiments were conducted as described in (A-B) except infection was carried out at an MOI of 10 PFU/cell comparing two different lab strains (KOS and 17syn+) and cell types (MRC-5 and Vero cells). All data points represent independent biological replicates and error bars represent logarithmic standard deviations. One-way ANOVA with a Dunnett’s multiple comparisons test was performed to compare differences between inhibited and uninhibited groups (n.s. not significant, * p < 0.1, ** p < 0.05, *** p < 0.01).
Fig 4.
PCNA-I1 reversibly blocks HSV-1 DNA replication.
A) Effects of PCNA-I1 and T2AA on HSV-1 DNA replication. MRC-5 cells were uninhibited or supplemented with PCNA-I1 (2.5 μM) or T2AA (12.5 μM) and infected with strain KOS at an MOI 10 PFU/cell. Total DNA was collected every 2 hours for 12 hours and at 24 hours post infection (hpi). The number of viral and cellular genomes were determined by qPCR relative to standard curves generated from purified viral or human DNA and the number of viral genomes per cell were determined. All values represent the means of biological triplicate experiments and error bars represent standard deviations. One-way ANOVA with a Dunnett’s test was performed to compare each time point of inhibited groups to the corresponding uninhibited control. Only statistically significant differences are shown (* p < 0.1, ** p < 0.05, *** p < 0.01). B) Effect of the time of PCNA-I1 addition on viral DNA replication. MRC-5 cells were supplemented with 2.5 μM PCNA-I1 either 24 or 1 hour before infection (hbi), at the time of infection, or at 1, 2, 3, 4, 5, or 6 hpi. Total DNA was collected at 12 hpi. Viral and cellular genome number were determined by qPCR as in (A). All values represent the mean of biological duplicate experiments and error bars represent standard deviation. One-way ANOVA with a Dunnett’s test was performed (* p < 0.1, ** p < 0.05, *** p < 0.01). C) Analysis of the reversibility of PCNA-I1 inhibition. MRC-5 cells were infected at an MOI 10 PFU/cell with strain KOS and were either uninhibited or supplemented with 2.5 μM PCNA-I1 one hour before and during infection (+PCNA-I1). In another sample, PCNA-I1 was removed at 6 hpi, cells were washed three times with TBS, and normal growth medium was replaced for the remainder of infection (+PCNA-I1 removed at 6 hpi). For all samples, virus was collected at 24 hpi. Viral yield was measured by plaque assay in Vero cells. All values represent the means of biological duplicate experiments with standard deviations. One-way ANOVA with Tukey’s multiple comparisons test was performed (* p < 0.1, ** p < 0.05, *** p < 0.01). D) PCNA is recruited to viral DNA despite PCNA inhibitor treatment. Vero cells were infected at an MOI 10 PFU/cell with strain KOS and were treated with 2.5 μM PCNA-I1 or 12.5 μM T2AA one hour before and during infection. EdC was incorporated into replicating viral DNA between 4–8 hpi and cells were fixed at 8 hpi. EdC labeled DNA was covalently attached to Alexa Fluor 488 (green) and PCNA was detected by immunofluorescence (red). Scale bars, 10 μM.
Fig 5.
Effect of PCNA inhibitors on the temporal cascade of viral protein expression.
A/C) Western blots of whole-cell lysates collected from cells infected or mock infected in the presence of PCNA-I1 (2.5 μM), T2AA (12.5 μM), or no inhibitor. MRC-5 cells were infected with HSV-1 strain KOS and total proteins were collected at 2, 4, and 6 hpi. Blots were probed with antibodies as indicated on the right. All infections were carried out at an MOI of 10 PFU/cell and all inhibitor treated cells were treated 1 hour before and throughout infection. B/D) Average fold change in protein expression (+inhibitor/-inhibitor) was determined and error bars represent standard deviations from biological duplicate experiments. Before calculating fold change, band intensities were normalized by dividing by the GAPDH signal detected from the same sample. Unpaired, two-tailed student’s t-tests were performed to compare the normalized band intensities of inhibited (+ PCNA-I1 or + T2AA) to uninhibited groups for each time point and each protein (* p < 0.1, ** p < 0.05, *** p < 0.01).
Fig 6.
Effects of PCNA inhibitors on viral mRNA levels.
MRC-5 cells were infected at an MOI 10 PFU/cell and supplemented with either PCNA-I1 (2.5 μM), T2AA (12.5 μM), or no inhibitor for one hour before and during infection. At 6hpi, total RNA was isolated, reverse transcribed, and amplified by qPCR. The number of viral mRNA copies per μg of total RNA was determined relative to a standard curve for each viral gene. A) ICP4 is a representative immediate early gene. B) TK, C) UL30, D) ICP8, and E) UL2 are representative early genes. F) UL42 is a representative leaky late gene. G) gC, H) gB, and I) gD are representative late genes. One-way ANOVA with a Dunnett’s multiple comparisons test was performed to compare inhibited groups to the corresponding uninhibited control (n.s. no significant difference, * p < 0.1, ** p < 0.05, *** p < 0.01).
Fig 7.
PCNA inhibitors cause decreased infectious virus production.
Cells were supplemented with either A) PCNA-I1 or B) T2AA, or as a control, uninhibited. Inhibitors were added one hour before and throughout infection. One million MRC-5 cells were infected at an MOI 10 PFU/cell in the presence or absence of 2.5 μM PCNA-I1 or 12.5 μM T2AA. Virus was collected 24 hours later. PFU was measured via plaque assay on Vero cells (Fig 3). Viral genomes per PFU were quantified by isolating viral DNA from collected virus, followed by qPCR relative to a standard curve. Unpaired, two-way t-tests were performed to compare inhibited to uninhibited groups (* p < 0.1, ** p < 0.05, *** p < 0.01).
Fig 8.
The use of NSAF to compare protein abundance in replicate viral aniPOND data sets and between different infection conditions.
Comparison of the NSAF of viral and cellular proteins that associate with viral genomes in viral aniPOND assays at 6 hpi. Conditions include A) KOS +EdC, B) KOS + 2.5 μM PCNA-I1 +EdC, or C) KOS + 12.5 μM T2AA +EdC. Each point represents an individual protein with the NSAF of that protein from one biological replicate plotted on the x-axis and another biological replicate plotted on the y-axis. The linear regression line is shown with the slope. The r value represents the calculated Pearson correlation coefficient. (D) The average NSAF of individual proteins was calculated and plotted for KOS +EdC (x-axis) and KOS +PCNA-I1 +EdC (y-axis) aniPOND datasets. (E) The graph was generated as in (D) except that the axes were adjusted to highlight proteins that are more or less abundant in the presence of PCNA-I1. Proteins that had at least 4-fold decreased association with PCNA-I1 are presented in Fig 9B. (F) and (G) Graphs were generated as in (D) and (E) except that the y-axis represents the average NSAF of individual proteins found in aniPOND datasets in the presence of T2AA. Red data points represent proteins that decreased at least 4-fold and green data points increased at least 4-fold in the presence of either PCNA-I1 or T2AA.
Fig 9.
Viral and cellular proteins associated with replicating HSV-1 DNA change due to PCNA inhibition.
A) Heat map indicating all viral proteins identified in viral aniPOND datasets and their fold change in abundance comparing PCNA-I1 (2.5 μM) or T2AA (12.5 μM) treated cells to an uninhibited control. HSV-1 genes are classified by their gene class (immediate early, early, leaky late, or late). Viral genes that are not classified are under “N/A”. Fold decrease was calculated as the NSAF of KOS+EdC divided by KOS +inhibitor +PCNA-1. Raw values indicate fold change and this is emphasized by the heat map. Values of less than 1.0 indicate an increase in protein association during inhibition. B) STRING diagram of human proteins that had at least a 4-fold decrease in association with viral DNA in the presence of 2.5 μM PCNA-I1. The STRING diagram shows predicted physical and functional interactions between human proteins. Biological process associated with identified proteins are labeled in varying colors. The unmapped list represents proteins that were not predicted to have high confidence protein-protein interactions with the other identified proteins using STRING. Some minor modifications were made to the STRING diagram to group known complex members together. ARID1A was originally grouped with ‘Mismatch repair proteins’ but was changed to be grouped with ‘Chromatin + DNA modification + remodeling.’ ERCC3 was originally grouped with the ‘Mediator complex’ but was changed to be grouped with ‘Transcription II’.
Fig 10.
MRN complex members associate with replicating viral DNA during PCNA-I1 inhibition.
Vero cells were infected at an MOI 10 PFU/cell with strain KOS and were either uninhibited or treated with 2.5 μM PCNA-I1 one hour before and during infection. EdC was incorporated into replicating viral DNA between 4–6 hpi and cells were fixed at 6 hpi. EdC labeled DNA was covalently attached to Alexa Fluor 488 (green) and A) Rad50 (GeneTex 13B3) or B) Mre11 (GeneTex 12D7) were detected by immunofluorescence (red). Scale bars, 10 μM. All images were taken using the same laser intensities. Traces were generated using the RGB profiler plugin in ImageJ and correspond to the white line drawn on the red/green merge (Merge (RG)) panel.
Fig 11.
Model depicting the effects of PCNA inhibitors on HSV-1 DNA replication and repair.
A) Model depicting proteins that associate with HSV-1 replication forks. PCNA is depicted in pink, UL30 in yellow, UL42 in purple, and the RNA primer in blue. Viral proteins are labeled to the left and cellular proteins are labeled on the right. B) T2AA is known to block PCNA protein-protein interactions via the PCNA IDCL (Fig 1D). Based on the data presented in this study, T2AA inhibition causes a defect in recruitment of UL2 to viral DNA. A PIP-like motif sequence is present in UL2, which is conserved in both the HSV-2 (UL2) and Epstein-Barr virus (EBV) (BKRF3) uracil DNA glycosylases (UNG). C) Proteins that are absent or significantly decreased on replicating viral DNA in the presence of PCNA-I1 are indicated with a red x or blue arrow, respectively. We present two models to explain these observations. D) Model 1: PCNA is unable to facilitate UL30 processivity and progression. E) Model 2 –During the second round of viral DNA replication, PCNA is unable to unload from the previously synthesized lagging DNA strand.