In Vivo Imaging of Influenza Virus Infection in Immunized Mice

Abstract

Immunization is the cornerstone of seasonal influenza control and represents an important component of pandemic preparedness strategies. Using a bioluminescent reporter virus, we demonstrate the application of noninvasive in vivo imaging system (IVIS) technology to evaluate the preclinical efficacy of candidate vaccines and immunotherapy in a mouse model of influenza. Sequential imaging revealed distinct spatiotemporal kinetics of bioluminescence in groups of mice passively or actively immunized by various strategies that accelerated the clearance of the challenge virus at different rates and by distinct mechanisms. Imaging findings were consistent with conclusions derived from virus titers in the lungs and, notably, were more informative than conventional efficacy endpoints in some cases. Our findings demonstrate the reliability of IVIS as a qualitative approach to support preclinical evaluation of candidate medical countermeasures for influenza in mice.IMPORTANCE Influenza A viruses remain a persistent threat to public health. Vaccination and immunotherapy are effective countermeasures for the control of influenza but must contend with antigenic drift and the risk of resistance to antivirals. Traditional preclinical efficacy studies for novel vaccine and pharmaceutical candidates can be time-consuming and expensive and are inherently limited in scope. In vivo imaging approaches offer the potential to noninvasively track virus replication in real time in animal models. In this study, we demonstrate the utility of bioluminescent imaging for tracking influenza virus replication in the lungs of immunized mice and also identify important factors that may influence the accurate interpretation of imaging results. Our findings support the potential of IVIS approaches to enhance traditional preclinical efficacy evaluation of candidate vaccines and human monoclonal antibodies for the prevention and treatment of influenza.

FIG 1

H1N1pdm09-NLuc is not attenuated in vitro or in BALB/c mice. MDCK cells were inoculated in triplicate at an MOI of 0.01 with either RG wt H1N1pmd09 or RG H1N1pdm09-NLuc. Supernatant was collected at the time points indicated for virus titration (A) or quantification of the luminescent signal (RLU, relative light units) (B). The correlation between supernatant luminescence and infectious virus titers (C) of H1N1pdm09-NLuc were estimated by Spearman’s rank order correlation (r = 0.72 [95% CI, 0.23 to 0.92]; P = 0.01). (D, E) Weight loss (mean ± standard deviation) (D) and survival (E) were analyzed in BALB/c mice (five per group) inoculated with either H1N1pdm09-NLuc or RG wt H1N1pdm09 virus. (F) In a separate experiment, mice (four per group) were challenged with 10³ TCID₅₀ of either H1N1pdm09-NLuc or RG wt virus. Infectious virus titers in lung homogenates were determined at the time points indicated. Dashed lines indicate the limits of detection of infectious virus (A, C, F) and the bioluminescent signal (B, C) or the threshold for euthanasia due to weight loss associated with infection (D).

FIG 2

Bioluminescent in vivo imaging of H1N1pdm09-NLuc replication in naive mice. Adult BALB/c mice were inoculated intranasally with 10³ TCID₅₀ of H1N1pmd09-NLuc virus. (A) A representative image series of a single mouse imaged at the time points indicated. A separate control mouse that was inoculated with 10³ TCID₅₀ of H1N1pmd09 RG wt virus and imaged at 7 dpi is shown at the far right for comparison. (B) Bioluminescence kinetics (mean ± standard error of the mean) are shown for the course of sublethal infection (n = 4 mice) and expressed in photons per second per square centimeter per steradian (p/s/cm²/sr). (C) Cohorts of mice four per group) were imaged and then immediately euthanized at several time points. The infectious virus titer in the lungs was plotted against the bioluminescent signal for each animal. The time points represented are 3, 5, 7, 10, and 12 dpi. The dashed horizontal line indicates the limit of detection of the bioluminescent signal. (D) Mice were challenged with either a sublethal (10³ TCID₅₀) or a lethal (10⁶ TCID₅₀) inoculum of H1N1pdm09-NLuc virus and imaged at the time points indicated.

FIG 3

Bioluminescent in vivo imaging of vaccinated mice. Adult BALB/c mice were inoculated intranasally with a single dose of homologous or heterologous LAIV or mock vaccinated with L-15. All mice were challenged with 10⁶ TCID₅₀ of H1N1pmd09-NLuc virus. Weight loss (A, D), a bioluminescence imaging series of a single representative mouse from each group (B, E), and virus titers in the lungs (C, F) are shown for homologous and heterosubtypic LAIV groups, respectively. The bioluminescent signal is expressed in photons per second per square centimeter per steradian (p/s/cm²/sr). (G) Cohorts of mice (n = 4 and n = 2 for the heterologous LAIV and mock-vaccinated groups, respectively) were imaged and then euthanized at 2, 4, and 6 dpi. Infectious virus titers (mean ± standard deviation) in the lung were plotted together with bioluminescent signal levels (mean flux in photons per second [p/s]). Hematoxylin and eosin staining (H) and immunostaining (I) of lung tissue sections collected at the times indicated were performed. Dashed lines indicate the threshold for euthanasia due to weight loss associated with infection (A, D) or the limit of detection of infectious virus (C, F, G). *, P < 0.05.

FIG 4

Bioluminescence in vivo imaging of passively immunized mice. An influenza virus-specific hMAb (EM4CO4) was administered either prophylactically (24 h prechallenge) or therapeutically (72 h postchallenge). Mice that received prophylaxis with an equivalent dose of isotype control antibody (human IgG1κ) are shown for comparison. (A to C) Weight loss (A); a bioluminescence imaging series of a single representative mouse from each group, expressed in photons per second per square centimeter per steradian (p/s/cm²/sr) (B); and virus titers in the lungs (C) are shown. (D) The mean bioluminescence in each group (four mice per group) is plotted for each time point indicated, and the area under the curve for each group is shaded to highlight differences in flux kinetics (photons per second). Magenta, prophylaxis; orange, therapy; gray, isotype control prophylaxis. Shaded arrows indicate the timing of antibody administration for each group that received influenza virus-specific antibody. Dashed lines indicate the threshold for euthanasia due to weight loss associated with infection (A), the limit of detection of infectious virus (C), or the limit of detection of the bioluminescent signal (D). *, P < 0.05.

FIG 5

In vivo bioluminescence imaging of mice receiving passive immunotherapy with an HA head versus stem antibody. The levels of protection conferred by hMAbs, one directed to the HA head (EM4CO4), and one directed to the HA stem (70-1F02), were compared (isotype control, IgG1κ, included for comparison). (A to C) Weight loss (A), virus titers in the lungs (B), and kinetics of chest bioluminescence (C) are shown for the time points indicated (four per group) and are expressed in photons per second per square centimeter per steradian (p/s/cm²/sr). (D) A representative image series of a single mouse is shown for each group. (E) Cohorts of mice (four per group) were imaged and then immediately euthanized at 2, 4, and 6 dpi. Infectious virus titers in the lungs (mean ± standard deviation) were determined and plotted together with bioluminescent signal levels expressed in photons per second (p/s). (F, G) Hematoxylin and eosin staining (F) and immunostaining (G) of lung tissue sections collected at the time points indicated were performed. Dashed lines indicate the threshold for euthanasia due to weight loss associated with infection (A), the limit of detection of infectious virus (B, E), or the limit of detection of the bioluminescent signal (C). *, P < 0.05.