Fig 1.
(A) SOSV-MARV experimental cohort and (B) KASV-MARV experimental cohort. DPI: Days post infection (relative to MARV inoculation). X: Procedure not performed.
Fig 2.
SOSV loads in specimens obtained from bats in the SOSV-MARV experimental cohort.
RT-qPCR-derived log10TCID50eq ml-1 SOSV loads in (A) blood, (B) oral swabs, and (C) rectal swabs. Days post MARV infection is shown for reference only. Symbols in A-C represent individual bats, and solid lines represent arithmetic mean viral loads. ND: Not detected.
Fig 3.
Pre- and post-study virus-specific IgG responses for bats in the SOSV-MARV and KASV-MARV experimental cohorts.
(A) Anti-SOSV IgG responses for bats in the SOSV-MARV experimental cohort, (B) anti-MARV nucleoprotein (NP) IgG responses for bats in the SOSV-MARV experimental cohort, (C) anti-KASV IgG responses for bats in the KASV-MARV experimental cohort, and (D) anti-MARV NP IgG responses for bats in the KASV-MARV experimental cohort. Post-study whole blood specimens were collected on the same day post MARV inoculation for all bats in the SOSV-MARV experimental cohort (day 18) and for all bats in the KASV-MARV experimental cohort (day 21). Symbols in A-D represent individual bats, solid lines represent arithmetic means, and dotted horizontal lines represent the cut-off values of the serological assays. Post-study whole blood specimens were collected on different days post SOSV infection for bats in A and different days post KASV infection for bats in C, and are therefore not comparable. The Shapiro-Wilks test determined that post-study anti-MARV NP IgG response datasets for coinfected and monoinfected bats in B and D are normally-distributed. Therefore, unpaired t-tests were used to determine if mean anti-MARV NP IgG significantly differed (P<0.05) between coinfected bats (n=12) and monoinfected bats (n=12). P values are located above each comparison and statistically significant values are bolded.
Fig 4.
MARV shedding dynamics in bats from the SOSV-MARV experimental cohort.
RT-qPCR-derived log10TCID50eq ml-1 MARV loads in (A) blood, (B) oral swabs, and (C) rectal swabs. Distribution of (D) peak oral MARV shedding loads, (E) the duration of oral MARV shedding, (F) peak MARV rectal shedding loads, and (G) the duration of rectal MARV shedding. Symbols in A-G represent individual bats, and solid lines represent arithmetic means. The Shapiro-Wilks test determined that the datasets represented in D, F, and G do not follow a normal or lognormal distribution, and the dataset represented in E is normally distributed. Therefore, Mann-Whiney U tests were used to determine if peak shedding loads (D and F) and the duration of shedding (G) significantly differed (P<0.05) between coinfected bats (n=12) and monoinfected bats (n=12), while an unpaired t-test was used to determine if the duration of shedding significantly differed between coinfected and monoinfected bats. P values are located above each comparison and statistically significant values are bolded. ND: Not detected.
Fig 5.
KASV loads in specimens obtained from bats in the KASV-MARV experimental cohort.
RT-qPCR-derived log10TCID50eq ml-1 SOSV loads in (A) blood, (B) oral swabs, and (C) rectal swabs. Days post MARV infection is shown for reference only. Symbols in A-C represent individual bats, and solid lines represent arithmetic mean viral loads. ND: Not detected.
Fig 6.
MARV shedding dynamics in bats from the KASV-MARV experimental cohort.
RT-qPCR-derived log10TCID50eq ml-1 MARV loads in (A) blood, (B) oral swabs, and (C) rectal swabs. Distribution of (D) peak MARV viremias, (E) the duration of MARV viremia, (F) peak oral MARV shedding loads, (G) the duration of oral MARV shedding, (H) peak MARV rectal shedding loads, and (I) the duration of rectal MARV shedding. Symbols in A-I represent individual bats, solid lines represent arithmetic means, and dashed lines in D represent geometric means. The Shapiro-Wilks test determined that the dataset represented in D is lognormally distributed, the dataset represented in G is normally-distributed, and the datasets represented in E, F, H, and I do not follow a normal or lognormal distribution. Therefore, to determine if peak shedding and the duration of shedding significantly differed (P<0.05) between coinfected (n=12) and monoinfected (n=12) bats, an unpaired t-test was applied to the log-transformed dataset represented in D, an unpaired t-test was applied to the dataset represented in G, and Mann-Whiney U tests were applied to the datasets represented in E, F, H, and I. P values are located above each comparison and statistically significant values are bolded. ND: Not detected.
Fig 7.
Cumulative MARV shedding in bats from the SOSV-MARV and KASV-MARV experimental cohort.
(A) RT-qPCR-derived log10TCID50eq ml-1 MARV cumulative shedding loads (sum of viral loads detected in oral swabs and rectal swabs from 0-18 days post MARV inoculation) for bats from the SOSV-MARV experimental cohort are denoted by symbols (solid lines represent arithmetic means) and represented on the left-y axis, and the percent of total cumulative shedding according to group are denoted by bars and represented on the right-y axis. (B) Lorenz curve of cumulative percentage of the SOSV-MARV experimental cohort bat population versus cumulative percentage of MARV shedding ranked in descending order (e.g., first symbol on the curve represents the bat which had the highest cumulative percentage of MARV shedding). (C) RT-qPCR-derived log10TCID50eq ml-1 MARV cumulative shedding loads (sum of viral loads detected in blood, oral swabs, and rectal swabs from 0-21 days post MARV inoculation) for bats from the KASV-MARV experimental cohort are denoted by symbols (dotted lines represent geometric means) and represented on the left-y axis, and the percent of total cumulative shedding according to group are represented by bars and shown on the right-y axis. (D) Lorenz curve of cumulative percentage of the KASV-MARV experimental cohort bat population versus cumulative percentage of MARV shedding ranked in descending order. The Shapiro-Wilks test determined that the cumulative shedding load dataset represented in A does not follow a normal or lognormal distribution and the cumulative shedding load dataset in C is lognormally-distributed. Therefore, to determine if cumulative shedding loads significantly differed (P<0.05) between coinfected (n=12) and monoinfected (n=12) bats, a Mann-Whiney U test was applied to the dataset represented in A and an unpaired t-test was applied to the log-transformed dataset represented in C. P values are located above each comparison and statistically significant values are bolded. ND: Not detected.