Polyamine Analog Diethylnorspermidine Restricts Coxsackievirus B3 and Is Overcome by 2A Protease Mutation In Vitro

Figures

• 

  Figure 1. DENSpm induces polyamine depletion and limits Coxsackievirus B3 infection. (A) Schematic of the mammalian polyamine biosynthetic pathway. Pertinent inhibitors are highlighted in green, while important enzymes are highlighted in lilac. (B) Vero-E6 cells were treated with 10, 50, and 100 μM DENSpm for 16 h prior to collection for Western blot for SAT1, above, and thin layer chromatography for polyamines measuring the presence of putrescine (Put), spermidine (Spm), and spermine (Spm), below. (C) Cells were left untreated (NT) or treated with 100 μM DENSpm for 16h prior to infection at multiplicity of infection of 0.01 CVB3. Viral titers were determined by plaque assay at the times indicated. (D) Cells were treated with escalating doses of DENSpm for 16h prior to infection at MOI 0.01. Viral titers were determined at 48 hpi. ** p ≤ 0.01using Student’s t-test (n ≥ 3), comparing treated samples to untreated controls. Error bars represent ± 1 SEM.

• 

  Figure 2. Coxsackievirus B3 gains resistance to DENSpm via mutation in 2A protease. (A) Vero-E6 cells were left untreated or treated with 100 μM DENSpm for 16 h prior to infection with CVB3 at an MOI 0.1. Virus was collected at 24 hpi and used to inoculate the next passage. Viral titers were determined via plaque assay for the passages shown. (B) CVB3 passaged 10 times over Vero-E6 cells, either treated with 100 μM or untreated, were used to infect Vero cells treated with increasing doses of DENspm for 24 hpi. Viral titers were determined by plaque assay. (C) Partial 2A protease sequence of CVB3 compared to other enteroviruses. Red arrows indicate amino acid residues of the protease active site, and the black arrow indicates mutated amino acid residue of 2AS35 mutants. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001 using Student’s t-test (n ≥ 3), comparing treated samples to untreated controls. Error bars represent ± 1 SEM.

• 

  Figure 3. CVB3 2AS35G confers resistance to DENSpm. (A) Vero cells were left untreated and were infected with CVB3 and 2A mutant protease at an MOI of 0.1. Samples were collected every 24 h and titered via plaque assay. (B) Vero cells were treated with increasing doses of DENSpm, from 10 μM to 100 μM, for 16 h prior to infection with wildtype CVB3 or 2A protease mutant. Viral titers were determined via plaque assay at 48 hpi. Viral titers were used to calculate the percent replication in DENSpm, by dividing the titer of the virus after infection of DENSpm-treated cells by the titer of the virus after infection of untreated cells at 48 hpi. (C) Thin layer chromatograms resolving the polyamines putrescine (Put), spermidine (Spm), and spermine (Spm) after treatment with DENSpm. Arrows indicate the band corresponding to each polyamine. (D) Vero cells were treated with increasing doses of DFMO, from 100 μM to 1 mM, for 4 days prior to infection with wildtype CVB3 or 2A protease mutant. Viral titers were determined via plaque assay at 48 hpi. Viral titers were used to calculate the percent replication in DFMO, by dividing the titer of the virus after infection of DFMO-treated cells by the titer…

• 

  Figure 4. CBV3 2AS35G exhibits reduced fitness in vitro. Veros cells were treated with 100 μM for 16 h or left untreated prior to being infected with wildtype and mutant CVB3 for 24 h, at which time (A) viral titers were determined via plaque assay. (B) WT and mutant CVB3 were used to infect untreated Vero cells for five passages at which time viral RNA was reverse transcribed, PCR amplified, and sequenced to measure stability and fitness of mutant. (C) Plaque sizes of virus mutants were determined after two-day plaque assay. (D) Pictures of plaques that were measured in (C). (E) Viral genomes in cell supernatant collected from untreated cells in (A) were determined by qPCR. The ratio of genomes-to-PFU was calculated by dividing the relative genomes by the titer in (A). * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001 using Student’s t-test (n ≥ 3), comparing treated samples to untreated controls. Error bars represent ± 1 SEM.

• 

  Figure 5. DENSpm reduces 2A protease activity. Split luciferase protease activity reporter systems were cloned and co-transfected with (A) wildtype 2A or (B) 2AS35G mutant into Vero cells left untreated or treated with increasing doses of DENSpm. Firefly luciferase activity was measured 24 h later and normalized to renilla luciferase transfection efficiency control and subsequently normalized to untreated cell transfection. (C) Vero cells were left untreated or treated with increasing doses of DENSpm for 16 h prior to infection with wildtype CVB3. Total cellular protein was collected 24 hpi and analyzed via Western blot for eIF4G and β-actin. ** p ≤ 0.01, *** p ≤ 0.001 using Student’s t-test (n = 2), comparing treated samples to untreated controls. Error bars represent ± 1 SEM.

• 

  Figure 6. Working model. (A) CVB3 relies on polyamines for efficient 2A protease activity to promote viral replication. In the presence of an inhibitor of polyamines, like DENSpm (B), protease activity and viral replication are reduced. (C) CVB3 overcomes polyamine depletion over successive rounds of replication via mutation of the 2A protease to facilitate rescued virus replication.