Evidence for Persistence of Ectromelia Virus in Inbred Mice, Recrudescence Following Immunosuppression and Transmission to Naïve Mice

Abstract

Orthopoxviruses (OPV), including variola, vaccinia, monkeypox, cowpox and ectromelia viruses cause acute infections in their hosts. With the exception of variola virus (VARV), the etiological agent of smallpox, other OPV have been reported to persist in a variety of animal species following natural or experimental infection. Despite the implications and significance for the ecology and epidemiology of diseases these viruses cause, those reports have never been thoroughly investigated. We used the mouse pathogen ectromelia virus (ECTV), the agent of mousepox and a close relative of VARV to investigate virus persistence in inbred mice. We provide evidence that ECTV causes a persistent infection in some susceptible strains of mice in which low levels of virus genomes were detected in various tissues late in infection. The bone marrow (BM) and blood appeared to be key sites of persistence. Contemporaneous with virus persistence, antiviral CD8 T cell responses were demonstrable over the entire 25-week study period, with a change in the immunodominance hierarchy evident during the first 3 weeks. Some virus-encoded host response modifiers were found to modulate virus persistence whereas host genes encoded by the NKC and MHC class I reduced the potential for persistence. When susceptible strains of mice that had apparently recovered from infection were subjected to sustained immunosuppression with cyclophosphamide (CTX), animals succumbed to mousepox with high titers of infectious virus in various organs. CTX treated index mice transmitted virus to, and caused disease in, co-housed naïve mice. The most surprising but significant finding was that immunosuppression of disease-resistant C57BL/6 mice several weeks after recovery from primary infection generated high titers of virus in multiple tissues. Resistant mice showed no evidence of a persistent infection. This is the strongest evidence that ECTV can persist in inbred mice, regardless of their resistance status.

Fig 1. Presence of ECTV-IFN-γbp^Δ in mice that survive infection.

(A) Survival proportions of BALB/c mice infected with 500 PFU ECTV-WT (n = 27) or ECTV-IFN-γbp^Δ (n = 58). P<0.0001, Kaplan-Meir log rank statistical test. Survival data combined from 4 separate experiments, with 6–15 mice per experiment. (B) ECTV-IFN-γbp^Δ titers in the indicated organs of BALB/c mice at days 7 and 21 p.i. (C) Lung and (D) spleen titers of ECTV-WT and ECTV-IFN-γbp^Δ. *, P<0.05; **, P<0.01 in comparing ECTV-WT and ECTV-IFN-γbp^Δ titers at the days p.i. Data shown in (B), (C) and (D) are composite viral loads determined in individual animals from 3 different experiments with groups of 3–6 mice and expressed as means of log₁₀ PFU/gram tissue ± SEM. The limit of virus detection is 2 log₁₀ PFU, shown by the dotted line. (E) ECTV genomes and (F) infectious virus in organs of BALB/c mice 37 days p.i. with ECTV-IFN-γbp^Δ. (G) ECTV genomes in organs of CBA/H mice 37 days p.i. with 500 PFU of ECTV-IFN-γbp^Δ. (H) ECTV genomes in organs of C57BL/6 mice 37 days p.i. with 500 PFU of ECTV-IFN-γbp^Δ. For E, F, G and H, n = 5 mice. The limit of virus genome detection is 10 copies, shown by the dotted line. BM = bone marrow; LN = popliteal lymph node; AG = adrenal gland.

Fig 2. CD8 T cell responses during acute ECTV infection.

Groups of female BALB/c mice were infected with 500 PFU ECTV-IFN-γbp^Δ, 500 PFU ECTV-WT or 2 x 10⁶ PFU ECTV-TK^Δ s.c., sacrificed on day 7 p.i. and splenic antigen-specific CD8 T cell responses were measured. (A) Percent specific lysis of ECTV-infected, ⁵¹Cr-labelled P815 targets by splenocytes from infected mice. ***, P<0.001 in comparing % specific lysis of virus-infected target cells by splenocytes from ECTV-WT infected mice with splenocytes from ECTV-IFN-γbp^Δ- or ECTV-TK^Δ-infected mice at the indicated effector-to-target ratios. (B) Percent specific lysis of virus-infected (Total) or ECTV CD8 T cell determinant-pulsed, ⁵¹Cr-labelled P815 targets by splenocytes from ECTV-IFN-γbp^Δ-infected mice. **, P<0.01 and ***, P<0.001 in comparing % specific lysis of L^d-026 peptide-pulsed targets with D^d-043 or K^d-149.5 peptide-pulsed targets by splenocytes from ECTV-IFN-γbp^Δ-infected mice at the indicated effector-to-target ratios. (C) Percent ECTV peptide determinant-specific IFN-γ⁺ CD8 T cells in spleens of ECTV-IFN-γbp^Δ-infected mice. L^d-026-restricted responses were significantly higher (P<0.01) compared with D^d-043- or K^d-149.5-restricted responses. (D) Numbers of ECTV-specific IFN-γ⁺ CD8 T cells in spleens of ECTV-IFN-γbp^Δ-infected mice. ***, P<0.001 for comparisons between groups shown. (E) Numbers of peptide-MHC class I tetramer⁺ CD8 T cells in spleens of ECTV-IFN-γbp^Δ-infected mice. ***, P<0.001 for comparisons between groups shown. (F) TCR Vβ chain expression by splenic CD8 T cells from virus-infected mice. P values for all panels were obtained by Mann-Whitney U test.

Fig 3. Antigen-specific CD8 T-cell responses at days 14 and 21 p.i.

Groups of 5 female BALB/c mice were infected with 500 PFU of ECTV-IFN-γbp^Δ and their spleen cells used to measure CD8 T cell responses. (A) Ex vivo cytolytic activity of splenocytes from ECTV-IFN-γbp^Δ infected mice against virus-infected P815 target cells on days indicated p.i. (B) Ex vivo cytolytic activity of splenocytes obtained from ECTV-IFN-γbp^Δ infected mice 21 days p.i. against ECTV-infected (Total) or ECTV peptide determinant-pulsed P815 target cells. ***, P<0.001. (C) Percent of ECTV peptide determinant-specific IFN-γ⁺ CD8 T cells. **, P<0.01 and ***, P<0.001. (D) Numbers of ECTV peptide determinant-specific IFN-γ⁺ CD8 T cells. **, P<0.01 and ***, P<0.001. (E) Numbers of peptide-MHC class I tetramer⁺ CD8 T cells. **, P<0.01 and ****, P<0.0001. (F) Percent ECTV-specific (total) IFN-γ⁺ CD8 T cells. (G) Numbers of ECTV-specific (total) IFN-γ⁺ CD8 T cells. **, P<0.01. P values for all panels were obtained by Mann-Whitney U test.

Fig 4. Long-term persistence of ECTV-specific CD8 T cell responses correlates with persistence of virus genomes in the BM and blood.

For data shown in panels A-E, groups of 6 BALB/c mice were infected with 500 PFU ECTV-IFN-γbp^Δ and functional and phenotypic assays done on days 7 till day 177. (A) Percent specific lysis of ECTV-infected, ⁵¹Cr-labelled P815 targets by splenocytes from infected mice at 100:1 effector: target ratio. (B) Percent specific lysis of ECTV peptide determinant-pulsed, ⁵¹Cr-labelled P815 targets by splenocytes from infected mice at 100:1 effector: target ratio. (C) Kinetics of ECTV-specific (total) IFN-γ⁺ CD8 T cells numbers. (D) Kinetics of ECTV peptide determinant-specific IFN-γ⁺ CD8 T cell numbers. (E) Kinetics of peptide-MHC class I tetramer⁺ CD8 T cell numbers. (F) Kinetics of ECTV genome copy numbers in BM. (G) Kinetics of ECTV genome copy numbers in blood. The limit of virus genome detection is 10 copies, shown by the dotted line. Data shown in panels F and G are from the same experiment in which 5 mice infected with 500 PFU ECTV-IFN-γbp^Δ were sacrificed on the indicated days but qRT-PCR assay run on the same day.

Fig 5. The effect of virus-encoded HRM and host resistance loci on virus persistence.

Groups of 5 female BALB/c mice were infected with 100 PFU of WT, IFN-γbp^Δ, IFN-α/βbp^Δ, IL-18bp^Δ, SPI-2^Δ, IFN-γbp^Δ-IL-18bp^Δ (double mutant) IFN-γbp^Δ-IL-18bp^Δ-SPI-2^Δ (triple mutant) ECTV. Separate groups of mice were sacrificed on days 7 and 35 to measure viral load. (A) Viral load in the liver at day 7 p.i. (B) Virus genome copy numbers in the BM at day 35 p.i. (C) Viral genome copy numbers in the BM of BALB/c mice 35 days p.i. with varying doses of ECTV-IFN-α/βbp^Δ or of ECTV-TK^Δ. For (A), (B) and (C), *, P<0.05; **, P<0.01; ***, P<0.001; ****, P<0.0001. (D) Survival of WT (BALB/c, BALB/b) and congenic (BALB/c.Cmv1^r, BALB/b.Cmv1^r) mice infected with 500 PFU ECTV-WT. P values were obtained by Log-rank (Mantel-Cox) test: P<0.001 in comparing % survival of BALB/c with BALB/c.Cmv1^r; P<0.05 in comparing % survival of BALB/b with BALB/b.Cmv1^r; P<0.0001 in comparing % survival of BALB/c with BALB/b; P<0.05 in comparing % survival of BALB/c.Cmv1^r with BALB/b. (E) NK cell and (F) CTL responses at day 7 p.i. in WT and congenic mice infected with 100 PFU ECTV-WT. (G) Virus genome copy numbers in the BM at day 35 p.i. in WT and congenic mice infected with 100 PFU ECTV-WT. (H) Virus genome copy numbers in the BM at day 35 p.i. in WT and congenic mice infected with 10⁵ PFU of ECTV-IFN-α/βbp^Δ. The limit of virus detection in A is 2 log₁₀ PFU, shown by a dotted line. The limit of virus genome detection in B, C, G and H is 10 copies, shown by the dotted line. For panels A-C, G and H, P values were obtained by Mann-Whitney U test for the indicated comparisons: *, P<0.05, **, P<0.01, ***, P<0.001 and ****, P<0.0001.

Fig 6. Immunosuppression with CTX triggers ECTV replication.

(A) Experimental scheme. Groups of 5 WT BALB/c mice were infected with 100 PFU of ECTV-WT or ECTV-IFN-γbp^Δ. Group 1 was sacrificed 80 days p.i. to measure viral load in various tissues, Group 2 was left untreated and Group 3 was treated with CTX. Three weeks later (day 101 p.i.), CTX-treated and untreated mice were sacrificed to measure viral load in organs. (B) ECTV-WT and (C) ECTV-IFN-γbp^Δ titers in organs of untreated or CTX-treated mice at 101 days p.i. BM, bone marrow; AG, adrenal gland. (D) CBA/H mice (n = 5/group) were infected with 100 PFU of ECTV-WT and 80 days p.i., one group was treated with CTX and the second group was left untreated. Shown are ECTV-WT titers in organs of untreated or CTX-treated CBA/H mice at 101 days p.i. For panels B-D, P <0.01 by Mann-Whitney U test for differences between untreated and CTX treated groups. (E) BALB/b.Cmv1^r mice (n = 5) were infected with 10⁵ PFU of ECTV-IFN-α/βbp^Δ and 80 days later treated with CTX 3 times over 15 days and sacrificed at 21 days post commencement of treatment (101 days p.i). (F) C57BL/NCrl (n = 3) and (G) C57BL/6J (n = 4) were infected with 1000 PFU of ECTV-WT and 80 days p.i., treated with CTX 4 times over 20 days. (F) ECTV-WT titers in organs of CTX-treated C57BL/6NCrl mice at 28 days post commencement of treatment (108 days p.i.) (G) Survival of C57BL/6J mice post commencement of treatment with CTX. One mouse that died on day 28 had high titers of virus in liver and spleen whereas no infectivity was detected in the organs of the remaining 3 animals that were sacrificed 35 days post commencement of treatment with CTX.

Fig 7. ECTV is transmitted from virus-infected, CTX-treated mice to naïve, co-housed mice.

(A) Experimental scheme. Groups of BALB/c (index) mice were infected with 50 PFU ECTV-WT, rested for 80 days and treated with CTX or left untreated. (B) ECTV-WT titers in organs of untreated or CTX-treated index mice at 21 days post-commencement of treatment (101 days p.i.). (C) Viral load in organs of individual CTX-treated mice at 101 days p.i. (D) ECTV-WT titers in organs of naïve mice co-housed with untreated or CTX-treated mice. (E) Viral load in organs of individual naïve mice 4 days post-separation from index mice. For C and E, the boxed data shows viral load in organs of mice that succumbed to mousepox.