Figure 1.
JCVCPN replicates viral DNA, but at lower levels than JCVMad-1.
(A) JCVCPN RR is archetype-like (adapted from (17)). (B) JCVCPN1.2 (hereafter referred to as simply JCVCPN) agnogene contains a 143 base pair deletion followed by a 75 base pair duplication at the beginning of the VP2 gene. (C–H) Cos-7, SVG and IMR-32 cells were transfected with linearized JCV genomes, or mock transfected. Cells were subcultured every 3–4 days and cell and supernatant samples were collected. DNA was extracted from the samples, digested with DpnI to remove input plasmid DNA, and analyzed by QPCR. Data represents the average of 4–10 independent experiments. In Cos-7 cell lysate (C) and supernatant (D) JCVMad-1 but not JCVCPN establishes a persistent infection. In SVG cell lysate (E) and supernatant (F) JCVMad-1 and JCVCPN both establish persistent infections. In IMR-32 cells, JCVMad-1 infection persists for 21 days, while JCVCPN becomes undetectable in cell lysate (G) and in supernatant (H). Error bars represent standard deviation. P-values were calculated using the Wilcoxon Rank Test. ND is not detected.
Figure 2.
JCVCPN expresses less early and late mRNA and VP1 protein than JCVMad-1.
(A–F) Cos-7, SVG and IMR-32 cells were transfected with JCVMad-1, JCVCPN or mock transfected and samples collected as described in Figure 1. qRT-PCR was used to determine the levels of early (T Ag) and late (VP1) transcripts. Relative Quantity (RQ) was calculated using the ΔΔCt method, using TATA-Box Binding Protein (TBP) as the endogenous control and JCVMad-1 as the calibrator sample. Data represents the average of 5–6 independent experiments. (A and D) In Cos-7 cells, JCVCPN expresses significantly lower levels of T Ag (A) and VP1 (D) mRNA. (B and E) In SVG cells, levels of T Ag (B) mRNA expressed by JCVCPN are similar to JCVMad-1, while VP1 (E) mRNA is significantly lower. (C and F) JCVCPN expresses significantly less T Ag (C) and VP1 (F) mRNA in IMR-32 cells. Error Bars represent standard deviation. P values were calculated for Student's t test using univariate analysis. ND is not detected. (G) Western blots were done with PAB597 (anti-VP1) using cell lysate from Cos-7 cells transfected with either JCVMad-1 (Lanes 1–3), JCVCPN (Lanes 7–9) or mock transfected cells (Lanes 4–6) collected 7 (Lanes1,4,7), 14 (Lanes 2,5,8) or 21 (Lanes 3,6,9) days post-transfection. VP1 can be detected in JCVMad-1 transfected cells at all time points, but not at any time with JCVCPN, with either a short (upper panel) or long (middle panel) exposure. Anti-tubulin antibody was used for loading control (lower panel). Blots are representative of 3 independent experiments (H) JCVMad-1, JCVCPN or mock transfected Cos-7 cells were analyzed for VP1 expression by flow cytometry. JCVMad-1 but not JCVCPN transfected cells have significantly higher levels of VP1 positive cells than Mock transfected samples. Results are the average of 4 independent experiments. Error bars represent standard deviation. P values were calculated using students t test, comparing JCVMad-1 and JCVCPN to mock.
Figure 3.
JCVCPN transfected IMR-32, SVG and Cos-7 cells produce low levels of infectious virions.
Supernatant from transfected Cos-7, SVG and IMR-32 cells was collected 7, 14 and 21 days post-transfection, and used to infect naive Cos-7 cells. At 7 days post-infection, cells were collected and either analyzed for JCV DNA using QPCR (A, C and D) or stained for VP1 and analyzed using flow cytometry (B). Supernatant collected 7 days post-transfection from JCVCPN transfected Cos-7 (A), SVG (C) or IMR-32 (D) cells can establish an infection in naïve Cos-7 cells, as measured by the presence of JCV DNA 7 days post-infection. Levels of DNA detected with JCVCPN infection are significantly lower than with JCVMad-1 infection (A, C and D). (B) VP1 positive cells are detected after infection with JCVMad-1, but not JCVCPN containing supernatant. Data is the average of 3–4 independent experiments. Error bars represent standard deviation. P vales were calculated using Wilcoxon rank test for QPCR data, comparing JCVCPN to JCVMad-1. P-values for flow cytometry data were calculated using students t test, comparing JCVMad-1 and JCVCPN to mock at each time point. ND is not detected.
Figure 4.
JCVMad-1 and JCVCPN chimeras and agno deletion mutants.
This diagram shows the various chimeric viruses and deletion mutants generated. Black represents sequences from JCVMad-1 and white represents sequences from JCVCPN. Lines represent DNA sequences. Dotted Lines represent deletions. Boxes represent genes. An X through ATG represents mutation of the start codon to eliminate protein expression. Mad-1 C-Agno is JCVMad-1 with the JCVCPN agnogene introduced. CPN M-Agno is JCVCPN with a full length agnogene from JCVMad-1 introduced. Mad-1 C-RR is JCVMad-1 with the JCVCPN RR and CPN M-RR is JCVCPN with the JCVMad-1 RR. Mad-1 Pt is JCVMad-1 with a mutated start codon which prevents the expression of the agnoprotein and Mad-1 Del is JCVMad-1 with the entire agnogene deleted. CPN M-Pt is JCVCPN with the agnogene of Mad-1 Pt.
Figure 5.
The agnogene deletion of JCVCPN is the primary cause of its replication defect.
Linearized JCV DNA of JCVMad-1, Mad-1 C-Agno, Mad-1 C-RR, Mad-1 Pt, Mad-1 Del, JCVCPN, CPN M-Agno, CPN M-RR, and CPN M-Pt were transfected into Cos-7 cells and DNA levels were quantified over 3 weeks, as described in Figure 1. (A and C) Levels of DNA with JCVMad-1 and the chimeras and agno deletion viruses on the JCVMad-1 backbone were measured in the cell lysate (A) and supernatant (C). Mad-1 C-Agno and Mad-1 Del show significantly lower levels of DNA late in infection. (B and D) Levels of DNA with JCVCPN and the chimeras and agno deletion viruses on the JCVCPN backbone were measured in the cell lysate (B) and supernatant (D). Introduction of a full length agnogene causes the greatest increase in DNA levels. Data represents the average of 4–10 independent experiments. Error bars represent standard deviation. P-values were calculated with the Wilcoxon Rank Test. ND is not detected.
Figure 6.
Deletion in the agnogene prevents VP1 expression.
Western blots for VP1 were done as described in Figure 2. (A) VP1 levels in Cos-7 cell lysate 14 days post-transfection. (B) VP1 levels in Cos-7 cells 21 days post-transfection. Levels of VP1 expression are drastically reduced in Mad-1 C-Agno, and expression is rescued by the full length agnogene in CPN M-Agno. Deletion of the agnogene Mad-1 Del results in a greater decrease in VP1 expression than just the prevention of the protein expression in Mad-1 Pt. Blots are representative of 3 or 4 independent experiments.
Figure 7.
Loss of the agnogene results in decreased production of infectious virions.
Supernatant was collected from Cos-7 cells 21 days post-transfection and used to infect naive Cos-7 cells. After 7 days, the cells were analyzed for (A) JCV DNA by QPCR or (B) VP1 expression by flow cytometry. Deletions in the agnogene in JCVMad-1-infected cells resulted in a significant decrease in JCV genomes/cell and percentage of cells expressing VP1, while introduction of a full length agnogene into JCVCPN results in a significant increase in the viral load and percentage of cells expressing VP1. Data is the average of 3 independent experiments. Error bars represent standard deviation. P values were calculated using Wilcoxon rank test for QPCR data, comparing each virus to its parent virus, either JCVMad-1 or JCVCPN. P-vales for flow cytometry data were calculated using students t test using the same sets of comparisons. ND is not detected.