Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, January 2005, p. 398-405, Vol. 49, No. 1
0066-4804/05/$08.00+0 doi:10.1128/AAC.49.1.398-405.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Department of Infectious Diseases,1 Animal Resources Center, St. Jude Children's Research Hospital, Memphis, Tennessee,2 BioCryst Pharmaceuticals, Inc., Birmingham, Alabama3
Received 12 April 2004/ Returned for modification 14 August 2004/ Accepted 22 September 2004
|
|
|---|
|
|
|---|
Interventions aimed at prevention or treatment of influenza virus or RSV infections have led to a reduction in secondary bacterial complications. Vaccination of children against influenza A virus was not only protective against influenza virus-induced illness but also decreased the incidence of bacterial otitis media (2, 6, 18). In vaccinated adults, a similar protective effect can be seen against secondary bacterial pneumonia (14, 36, 37, 38). Although no vaccines are currently licensed for use against RSV or hPIVs, passive immunization with a high dose of polyclonal RSV antibody preparation has been demonstrated to prevent otitis media (42). In cases in which prevention of preceding viral infection cannot be accomplished or has been inadequate, the use of antiviral drugs might be an appropriate strategy to prevent later bacterial complications. Among viral respiratory illnesses, effective inhibitors are licensed only for influenza viruses at present. Zanamivir and oseltamivir are two members of a novel class of potent and selective inhibitors of the influenza virus neuraminidase (NA) that have been licensed for treatment of influenza A and B virus infections in recent years (16). Although no studies primarily designed to determine the effect of these drugs on prevention of secondary complications of influenza in humans have been published, some effects can be seen even with this caveat. Early oseltamivir treatment reduced the development of acute otitis media by 44% in influenza-infected children 1 to 12 years old (50). In adults, early oseltamivir or zanamivir treatment of influenza reduced the occurrence of secondary complications (otitis media, sinusitis, bronchitis, or pneumonia), as well as antibiotic use (33, 47). Inhibitors of the human parainfluenza virus hemagglutinin-NA (HN) might be expected to have similar effects on secondary bacterial infections after infections with PIVs.
This is the first report describing the efficacy of the novel parainfluenza virus HN inhibitor BCX 2798 in the prevention of secondary Streptococcus pneumoniae infections in an animal model. BCX 2798 was designed and synthesized upon the available crystal structure of the HN protein of Newcastle disease virus (NDV) (9, 45). We previously reported the efficacy of this novel compound against parainfluenza viruses both in vitro and in vivo (1). BCX 2798 successfully inhibited binding and NA activities of hPIV-1, hPIV-2, and hPIV-3 viruses (with the highest level activity against the HN of hPIV-1) and was highly efficacious in prophylaxis of lethal infection in mice with a recombinant parainfluenza virus bearing the hPIV-1 HN (instead of the Sendai virus [SV] HN) on an SV background. This virus, denoted rSV(hHN), was used in a prior study (1) because the hPIVs infect experimental animals poorly, infection is usually asymptomatic, and lung pathology is minimal or undetectable (7, 29, 35, 46). In contrast, infection of mice with the recombinant virus rSV(hHN) causes severe illness and robust replication in the lungs of mice (1). In the present study we utilized this virus model to develop and characterize the synergism between PIVs and S. pneumoniae in mice. We hypothesized that the novel parainfluenza virus inhibitor BCX 2798 would prevent the synergistic mortality caused by rSV(hHN) and pneumococcus. The experiments demonstrate that rSV(hHN) predisposes for development of lethal pneumococcal pneumonia in mice, and the BCX 2798 compound prevents this lethal synergism in a prophylaxis treatment model.
|
|
|---|
![]() View larger version (11K): [in a new window] |
FIG. 1. Chemical structure of BCX 2798 and its parent compound. (A) Neu5Ac2en (2-deoxy-2,3-dehydro-N-acetylneuraminic acid); (B) BCX2798 (4-azido-5-isobutyrylamino-2,3-didehydro-2,3,4,5-tetradeoxy-D-glycero-D-galacto-2-nonulopyranosic acid).
|
Infectious agents. rSV(hHN) virus, which contains the hPIV-1 HN gene instead of the HN gene of SV, was rescued by reverse genetic methods. Briefly, SV (strain E) was rescued from the full-length SV cDNA genome pSeV(+) as described previously (26). The full-length cDNA clone of SV was mutated to include an NotI site upstream and an AscI site downstream of the HN gene, creating the pSV(+)AN plasmid. These restriction sites were also added to the hHN cDNA (from hPIV-1) and used to exchange the HN gene. For the rescue of the recombinant virus, 293T cells were infected with the vaccinia virus vTF7-3, which expresses T7 RNA polymerase, and transfected with the full-length rSV(hHN) genome, as well as the NP, P, and L genes in expression vectors as described previously (3). Two days after transfection, the infected and transfected cells were subjected to three cycles of freezing and thawing and injected into the allantoic cavity of 10-day-old embryonated chicken eggs to amplify the virus. The rescued virus was plaque purified on LLC-MK2 cells, amplified in embryonated chicken eggs, divided into aliquots, and kept frozen at 70°C. The sequence of the HN gene of egg-grown rSV(hHN) did not differ from that of wild-type hPIV-1. Infectivity of virus was determined by endpoint dilution assay in LLC-MK2 [to determine the dose(s) of virus in 1 ml that infects 50% of cells in culture (TCID50/ml)]. S. pneumoniae D39, a type 2 encapsulated strain kindly provided by Elaine Tuomanen (St. Jude Children's Research Hospital, Memphis, Tenn.), was grown in Todd-Hewitt broth (Difco Laboratories, Detroit, Mich.) and frozen at 70°C in aliquots of known concentration.
Drug studies with mice. Animal studies with 8- to 10-week-old female 129x1/SvJ mice (weight, 18 to 20 g; Jackson Laboratories, Bar Harbor, Maine) were performed in a biosafety level 3 facility in the Animal Resource Center at St. Jude Children's Research Hospital. All experiments were done under general anesthesia with inhaled isoflurane 2.5% (Baxter Healthcare Corp., Deerfield, Ill.) and were approved by the Animal Care and Use Committee at St. Jude Children's Research Hospital.
Mice were infected by intranasal inoculation with 50 µl of sterile phosphate-buffered saline (PBS) containing a sublethal dose of 105.5 TCID50 of rSV(hHN) per mouse and challenged 7 days later with 100 µl of sterile PBS containing a sublethal dose of 25 CFU of S. pneumoniae per mouse. Treatment with BCX 2798 was by intranasal administration in a volume of 50 µl of sterile PBS twice daily for 5 days starting 4 h before or 24 h after virus exposure. BCX 2798 compound was administered to groups of mice (n = 10) at dosages of 0.1, 1, 10, and 50 mg/kg per day. Control (infected untreated) animals received PBS instead of drug. Administration of drug or PBS was done under light anesthesia with isoflurane, and care was taken to ensure that exposure to anesthesia was equal for all groups. Mice were observed daily for 21 days after the pneumococcal challenge for clinical signs of infection and for survival. The parameters for evaluation of antiviral activity of BCX 2798 included prevention of weight loss and death and prolongation of survival. Weight changes were calculated for each mouse as a percentage of its weight on day 0 before virus infection.
Lung viral and bacterial titers. Four mice from each group were sacrificed at 4, 48, and 96 h after pneumococcal infection for virus and bacterium titrations. Lungs were removed under sterile conditions, washed three times in PBS, and ground and suspended in a total volume of 1 ml. The suspensions for virus titration were centrifuged at 2,000 x g for 10 min to clear cellular debris. Virus titers were determined by plating tenfold dilutions of this suspension in a final volume of 1 ml on LLC-MK2 cells in 24-well plates with minimal essential medium containing 0.1% bovine serum albumin in the presence of 1 µg of acetylated trypsin and 50 µg of gentamicin/ml to determine the TCID50. Lung homogenates were used directly for bacterial cultures prior to centrifugation. Quantitation of pneumococcal colony counts was done by using 10-fold dilutions on tryptic soy agar plates (Difco) supplemented with 3% (vol/vol) sheep erythrocytes.
Pathology. Lungs were removed on day 3 after the secondary challenge and fixed in 10% neutral buffered formalin for 24 h. Lungs were then embedded in paraffin, sectioned (5 µm), stained with hematoxylin and eosin, and examined microscopically for histopathologic changes by an experienced animal pathologist (K.L.B.) who was blinded to the composition of the study groups. The degree of involvement of lung sections was estimated by examination under low power for evidence of consolidation, whereas the type of involvement, including the specific cell types involved, was determined by examination under high power. The airways (bronchioles, alveolar ducts, etc.) were examined to see whether they contained inflammatory cells, epithelial necrosis, or epithelial hyperplasia.
Statistical analysis. Comparison of survival between groups of mice was done with the Mantel-Cox chi-square test on the Kaplan-Meier survival data. Comparison of viral and bacterial titers in lungs between groups was done with the Wilcoxon rank sum test. Comparisons of weight loss and mean day to death between groups were done by using the Student t test for pairwise comparisons and one-way analysis of variance, followed by Dunn's test for comparisons among multiple groups. The mean survival day was determined as the number of days surviving after infection considering only mice that died. A P value of <0.05 was considered significant for these comparisons.
|
|
|---|
To establish that lethal synergism occurs between a paramyxovirus pseudotyped with the hPIV-1 HN and pneumococcus, we first determined the maximum infectious dose at which no mortality occurred for rSV(hHN) and for S. pneumoniae in 129x1/SvJ mice (Table 1). The sublethal doses of 105.5 TCID50 of recombinant virus per mouse and of 25 CFU of pneumococci per mouse caused moderate illness but no mortality of infected animals and were used for further experiments.
|
View this table: [in a new window] |
TABLE 1. Survival of 129x1/SvJ mice infected with rSV(hHN) or S. pneumoniae
|
|
View this table: [in a new window] |
TABLE 2. Effect of sequence of rSV(hHN) and S. pneumoniae inoculation on survival
|
![]() View larger version (20K): [in a new window] |
FIG. 2. Effect of different doses of rSV(hHN) on weight loss (A) and survival (B) of mice in a secondary pneumonia model. Mice (n = 10 per group) were infected with a sublethal dose of 103.5, 104.5, or 105.5 TCID50 per mouse of rSV(hHN) and then challenged 7 days later with a sublethal dose of 25 CFU of S. pneumoniae per mouse. Mice in a control virus group were challenged with PBS instead of bacteria and had 100% of survival (data not shown). Weight loss and survival of mice were monitored through 21 days after the bacterial challenge. Mean values for weight loss are plotted with error bars indicating the standard deviation. An asterisk indicates a significant difference in weight change or survival compared to the control groups infected with virus or bacteria alone (P < 0.05).
|
Prophylactic treatment with BCX 2798 was administered to mice at doses of 0.1, 1, and 10 mg/kg per day in two divided doses, starting 4 h before viral infection.
Weight loss and survival of infected treated animals were compared to weight loss and survival of infected untreated animals and were monitored for 21 days after the bacterial challenge (Fig. 3). Mice infected with a sublethal dose of rSV(hHN) and challenged 7 days later with a sublethal dose of S. pneumoniae (placebo-treated synergism group) lost an average of 35% of their initial weight, and all died by day 11 after the secondary pneumococcal infection. Mice infected in the same manner but treated with BCX 2798 at 10 mg/kg per day starting 4 h before viral infection (BCX 2798 treated synergism group) lost an average of 3% of their initial weight by day 7 after the viral infection but did not lose further weight upon bacterial challenge. BCX 2798 at 10 mg/kg per day protected mice from mortality as well, significantly improving survival to 80% compared to 0% in the placebo-treated group. Pretreatment of mice with the lower doses of BCX 2798 of 0.1 and 1 mg/kg per day did not prevent weight loss of infected animals after the second bacterial challenge and had no effect on synergistic mortality (data not shown).
![]() View larger version (19K): [in a new window] |
FIG. 3. Effect of BCX 2798 on weight loss (A) and survival (B) of mice in a secondary bacterial pneumonia model after rSV(hHN) infection. Mice (n = 10 per group) were infected with a sublethal dose of 105.5 TCID50 of rSV(hHN) per mouse and then challenged 7 days later with a sublethal dose of 25 CFU of S. pneumoniae per mouse. Mice in the control groups were infected either with virus or with bacteria alone. Treatment at 10 mg/kg per day with BCX 2798 or PBS as a placebo was started 4 h before viral infection. Weight loss and survival of mice were monitored through 21 days after the bacterial challenge. Mean values for weight loss are plotted with error bars indicating the standard deviation. An asterisk indicates a significant difference in weight change or survival for the drug-treated group compared to the synergism group receiving placebo (P < 0.05).
|
Lung viral and bacterial titers.
To further evaluate the model of synergistic mortality and the effect of prophylactic treatment in the model, we repeated the experiment utilizing BCX 2798 administered to mice 4 h before viral infection at a dosage of 10 mg/kg per day and measured bacterial and viral titers in mouse lungs. Lung titers were assayed in groups of 4 mice at 4, 48, and 96 h after the secondary bacterial infection (Fig. 4). Virus was only detected in the lungs of mice at the 4 h time point after bacterial challenge (or day 7 after viral infection), a result consistent with our earlier data demonstrating the clearing of rSV(hHN) from mouse lungs at doses of less than 106 TCID50 before day 9 after viral infection (1). The drug-treated synergistic group of mice had 4 h postchallenge median viral lung titers ca. 2 logs lower than did placebo-treated mice or mice infected with recombinant virus alone (Fig. 4A). Bacterial superinfection did not increase virus load in mouse lungs and did not delay viral clearance. Lung bacterial titers were noted to increase significantly (ca. 4 logs) between the 4- and 48-h time points in both synergistic groups of mice but not in the control mice infected with S. pneumoniae alone (Fig. 4B). Bacterial titers in this last group were undetectable at all time points. Although we observed an increase in bacterial titers in both synergistic groups (BCX 2798-treated and placebo-treated mice) the bacterial titers in the placebo-treated group were significantly higher (
1 log) compared to the BCX 2798-treated group of mice. No titers were done in the untreated synergistic group of mice at 96 h after secondary challenge since all of them died before that time point. Thus, preceding parainfluenza virus infection led to an increase in lung bacterial titers, and protection of mice from lethal synergism by BCX 2798 was mediated by a reduction in lung bacterial titers as a result of inhibition of rSV(hHN) growth in mouse lungs.
![]() View larger version (15K): [in a new window] |
FIG. 4. Lung viral (A) and bacterial (B) titers in a secondary pneumonia model following rSV(hHN) infection. Mice (n = 4 per group) were infected with a sublethal dose of 105.5 TCID50 of rSV(hHN) per mouse and then challenged 7 days later with a sublethal dose of 25 CFU of S. pneumoniae per mouse. Mice in the control group were infected either with virus or bacteria alone. Treatment at 10 mg/kg per day with BCX 2798 or PBS was started 4 h before the viral infection in the treatment and placebo groups. Lungs were collected 4, 48, and 96 h after challenge with S. pneumoniae. Each point represents the results from a single mouse. A single asterisk indicates a significant difference in titers at that time point compared to the control group infected with a single agent (P < 0.05), and two asterisks indicate a significant difference in titers at that time point compared to the placebo group of mice. ND, not determined because of death of mice.
|
![]() View larger version (151K): [in a new window] |
FIG. 5. Histopathologic changes in the lungs of mice infected in a secondary bacterial pneumonia model. Mice received PBS only (A); S. pneumoniae D39 at 25 CFU per mouse 7 days after PBS mock infection (B); rSV(hHN) at 105.5 TCID50 per mouse, followed 7 days later by PBS mock infection (C); rSV(hHN), followed 7 days later by pneumococci with placebo treatment (D); or rSV(hHN), followed 7 days later by pneumococci with BCX 2978 treatment (E). High-power magnification (x40) views of representative lung sections obtained 3 days after secondary challenge and stained with hematoxylin and eosin are pictured. The arrow in panel D highlights an area of extensive necrosis that is not present in the lungs of mice represented by panel E.
|
|
|
|---|
Our experiments demonstrate that the recombinant virus rSV(hHN) can prime mice for the development of pneumococcal pneumonia. To achieve synergistic killing, virus infection must precede the bacterial challenge since preinfection with pneumococci did not increase morbidity or mortality from rSV(hHN), and simultaneous administration caused only mild enhancement of disease. The magnitude of the effect on mortality was related to the dose of virus administered, although doses below those needed to cause clinical symptoms in mice could prime for lethal infection. Thus, in some respects the model is similar to that developed for influenza virus and pneumococcus (32), reinforcing the concept that respiratory viruses other than influenza virus predispose to secondary bacterial infections.
The novel parainfluenza virus inhibitor BCX 2978 is a rationally designed compound based on the crystal structure of the HN of NDV. The drug binds the active site of the HN of hPIVs, inhibiting both hemagglutination and NA activity over a broad range of concentrations (1). Because both binding and NA activities are contained in a single site, inhibition of the hPIV-1 HN by competitive blockade likely has different effects on the virus life cycle than inhibitors of the influenza virus NA have on influenza virus. In earlier experiments, we demonstrated that inhibition of the influenza virus NA by prophylactic treatment with oseltamivir decreased viral replication and inhibited lethal synergism with pneumococcus in a mouse model, similar to the results we report here (29). This effect was specific to the NA and was thought to be mediated by cleavage of sialic acids by the viral NA during replication and spread of the virus throughout the respiratory tract, exposing receptors for pneumococcal adherence and invasion. However, a separate treatment effect was seen in the influenza virus model that was dependent on inhibition of NA activity but was independent of viral replication, implying other functions for the influenza virus NA in the context of a secondary infection. This effect of delayed treatment was not seen in the current study using rSV(hHN). The reasons for this are unclear. A reduced dependency on NA activity for viral spread is one potential explanation, since the hPIV-1 HN has relatively low NA activity compared to SV or influenza viruses. The ability of BCX 2978 to inhibit synergism when viral replication is decreased but not in a delayed treatment model when viral replication remains unchanged indicates that the contribution of the viral NA is less in this model than in the previous model utilizing influenza virus. The similar spectrum of synergistic interaction shared between the hPIVs and RSV, a virus that lacks NA activity, supports this argument. Since the HN switches between conformations suitable for binding or for hydrolysis, both of which take place at the same site (8, 9), NA activity may not be expressed at the time it would be needed for support of bacterial adherence. Thus, inhibition of the HN would have no effect on this proposed mechanism of synergism.
It is more likely that inhibition of viral spread through blockade of attachment at the HN site, or a combined effect on attachment and NA activity, is responsible for the reduction in secondary bacterial disease. The main effect seen in mice superinfected with pneumococci was an increase in the lung bacterial load. This increase was rapid over the first 48 h and resulted in severe consolidation from an inflammatory infiltrate throughout the lungs. The degree of consolidation throughout the lungs and the presence of necrosis were striking differences from the characteristic pneumococcal pneumonia seen with bacterial infection alone. Interestingly, the virus titers of rSV(hHN) were not increased in pneumococcal superinfected mice. This failure to augment viral replication is another difference compared to the influenza model (30) that may provide clues as to the mechanism. In the prior model, bacterial infection increased the lung influenza virus load by >10-fold, and prophylactic drug treatment prevented this bacterium-mediated increase (29). Although pneumococcus did not increase lung titers of rSV(hHN) in the absence of drug in the present study, treatment with BCX 2978 led to lower lung titers of both bacteria and virus. This resulted in a change in the distribution of lesions in the lungs but not in the character of the lesions. Necrosis and inflammatory infiltrates were present, but the pathological changes were limited in extent and more closely resembled early stages of infection. Thus, the effect of viral inhibition was to slow the progression of secondary bacterial pneumonia that likely allowed the immune system to clear the infection before it overwhelmed the animals. Although much recent attention has been focused on mechanisms by which paramyxoviruses suppress interferons (15, 27), it is not clear how this might benefit bacteria. It is not known whether these viruses have other effects on the innate and adaptive immune systems that might enhance bacterial growth.
The interaction between paramyxoviruses and bacteria is clinically significant. The hPIVs have a widespread distribution in the world and are one of the leading causes of hospitalization from respiratory virus infections in children (40). The hPIVs have a role in the development of bacterial otitis media and secondary bacterial pneumonia, and the present study underscores the importance of this interaction. Development of secondary bacterial complications as a result of primary PIV disease increases the economic impact of viral infection enormously. Our results suggest that an approach using selective inhibitors of the HN of PIVs might be reasonable not only for the amelioration of viral disease but also for prevention of secondary bacterial complications. Although our data in a mouse model support such a role for prophylaxis but not treatment, use of antiviral agents active against paramyxoviruses for this purpose should be studied in humans in both clinical settings. Intervention would be expected to be especially efficacious in children, the elderly population, immunosuppressed persons, and in patients with antibiotic treatment failure. These findings support further drug development and highlight the need for clinical trials with humans to address these issues.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»