Pneumocystis carinii Infection Causes Lung Lesions ...

Comparative Medicine Copyright 2011 by the American Association for Laboratory Animal Science

Vol 61, No 1 February 2011 Pages 45–52

Pneumocystis carinii Infection Causes Lung Lesions Historically Attributed to Rat Respiratory Virus Robert S Livingston,* Cynthia L Besch-Williford, Matthew H Myles, Craig L Franklin, Marcus J Crim, and Lela K Riley Idiopathic lung lesions characterized by dense perivascular cuffs of lymphocytes and a lymphohistiocytic interstitial pneumonia have been noted in research rats since the 1990s. Although the etiology of this disease has remained elusive, a putative viral etiology was suspected and the term ‘rat respiratory virus’ (RRV) has been used in reference to this disease agent. The purpose of this study was to determine whether Pneumocystis carinii infection in immunocompetent rats can cause idiopathic lung lesions previously attributed to RRV. In archived paraffin-embedded lungs (n = 43), a significant association was seen between idiopathic lung lesions and Pneumocystis DNA detected by PCR. In experimental studies, lung lesions of RRV developed in 9 of 10 CD rats 5 wk after intratracheal inoculation with P. carinii. No lung lesions developed in CD rats (n = 10) dosed with a 0.22-µm filtrate of the P. carinii inoculum, thus ruling out viral etiologies, or in sham-inoculated rats (n = 6). Moreover, 13 of 16 CD rats cohoused with immunosuppressed rats inoculated with P. carinii developed characteristic lung lesions from 3 to 7 wk after cohousing, whereas no lesions developed in rats cohoused with immunosuppressed sham-inoculated rats (n = 7). Both experimental infection studies revealed a statistically significant association between lung lesion development and exposure to P. carinii. These data strongly support the conclusion that P. carinii infection in rats causes lung lesions that previously have been attributed to RRV. Abbreviations: RRV, rat respiratory virus; PCP, Pneumocystis pneumonia.

In the late 1990s, a series of reports emerged describing inflammatory lesions in the lungs of rats used in research for which an etiology could not be identified.11,12,16,24,26 One of the first reports described lesions in male and female F344 rats on multiple prechronic toxicity studies performed over several years in different facilities in the United States.11 The lesions consisted of prominent increases in perivascular lymphocytes throughout the lung. Other features were infiltrates of macrophages, neutrophils, and lymphocytes within alveolar spaces, focal hyperplasia of type 2 pneumocytes, and a variable increase in peribronchiolar lymphoid tissue. In control rats of one study, lesions were seen in most rats at the 13 wk time point, whereas rats evaluated at 6 and 12 mo had a marked decrease in lesion incidence. Similar lung lesions were not seen in F344 rats evaluated at the end of 2-y toxicity studies. Similar lesions developed in male and female Wistar Hsd/Cpb:Wu rats used in toxicologic research.26 Over a 3-y time frame, lesions were seen in 50% of young Wistar rats used in 2- and 4-wk toxicity studies, with the incidence and severity of lesions decreasing by 13 and 26 wk and absent by 1 y. The authors speculated that resolution of the lung lesions may result from the acquisition of immunity, indirectly suggesting an infectious etiology for this condition.26 Recently, the transmission of idiopathic lung lesions from rats in a colony endemic for lung lesions to Wistar Han rats (Crl:WI[Han]) originating from a colony negative for lung lesions was demonstrated.1 The Wistar Han rats began to develop idiopathic lung lesions by 5 wk after exposure, and affected rats were detected weekly for the remainder of the 13-wk study. This same report1 proposed histologic criteria for Received: 03 Sep 2010. Revision requested: 04 Oct 2010. Accepted: 09 Oct 2010. Research Animal Diagnostic Laboratory, Columbia, Missouri. * Corresponding author. Email: [email protected]

the diagnosis of idiopathic lung lesions in rats to be “dense cuffs of lymphocytes, often with some plasma cells and macrophages around multiple blood vessels in multiple areas of the lung and/ or lymphohistiocytic interstitial pneumonia.” Increased peribronchiolar lymphoid tissue was not seen in diseased rats, and consistent with other reports, is not a diagnostic feature of this disease.4,24,25 This previous study,1 along with other reports, provides further suggestive evidence that the etiology of idiopathic lung lesions in rats is a transmissible infectious agent.1,4,24,25 To date, attempts to identify the etiology of idiopathic lung lesions in rats have been unsuccessful.4,24,25 However, in early investigations, a limited cytopathic effect was seen in cell lines incubated with lung homogenates from diseased rats, and lung lesions were reproduced in rats inoculated with the cultured cells.4,25 In conjunction with the perivascular and interstitial lesion distribution and lack of an identifiable etiologic agent, these data led researchers to suspect a viral etiology. The term ‘rat respiratory virus’ (RRV) was adopted to confer a putative viral etiology to the idiopathic lung lesions and has been used over the past decade in reference to this disease. Idiopathic interstitial pneumonia has a world-wide distribution and remains an important problem in rats used for biomedical research, in part because a causative agent has not been identified, and diagnostic tests other than histologic evaluation of lungs are not available.1,16,22,24-26 In a survey of rats from research institutions in North America, 6% of rats evaluated had histologic evidence of idiopathic lung lesions consistent with RRV.22 We identified several immunocompetent rats with histopathologic idiopathic lung lesions of RRV that also tested positive for Pneumocystis by PCR. Pneumocystis are single-celled fungal respiratory pathogens of mammals that are transmitted by the airborne route17,27 and known to cause a severe, often lethal, pneumonia in

45

CM10000095.indd 45

3/4/2011 1:34:26 PM

Vol 61, No 1 Comparative Medicine February 2011

immunocompromised hosts.8,28,29 Experimental infection studies have shown that Pneumocystis isolates are host species-specific, with interspecies transmission not occurring even in immunodeficient hosts.2,10,14,15 P. carinii and P. wakefieldiae are the 2 Pneumocystis species that have been identified in laboratory rats, with P. carinii being most commonly found.9,30 Widespread presence of P. carinii has been documented in the lungs of clinically healthy commercially produced immunocompetent rats.18,19 Neonatal rats acquire infection within hours after birth and can harbor the organism without evidence of clinical sequelae unless subjected to chronic immune suppression.18,19 The clinical or histopathologic consequences of subclinical P. carinii infections in immunocompetent rats have not been investigated. To investigate the association between Pneumocystis infection and lesions otherwise attributed to RRV, we conducted experiments to test the hypothesis that P. carinii infection in immunocompetent rats causes idiopathic lung lesions that have been previously attributed to the putative RRV. In addition, experiments were conducted to rule out a viral etiology for this disease.

Materials and Methods

Rats. Female 4-wk-old Crl:CD(SD) rats (Charles River, Wilmington, MA) were obtained from a colony that was historically negative for histologic lesions consistent with so-called RRV. In addition, health monitoring reports from the colony were negative for Sendai virus, pneumonia virus of mice, sialodacryoadenitis virus (rat coronavirus), Kilham rat virus, Toolan H1 virus, rat parvovirus, rat minute virus, reovirus, rat theilovirus, lymphocytic choriomeningitis virus, hantavirus, mouse adenovirus, Bordetella bronchiseptica, cilia-associated respiratory bacillus, Clostridium piliforme, Corynebacterium kutcheri, Mycoplasma pulmonis, Pasteurella multocida, Pasteurella pneumotropica, Pseudomonas aeruginosa, Salmonella spp., Streptococcus pneumoniae, Encephalitozoon cuniculi, ectoparasites, endoparasites, and enteric protozoa. Rats were housed in autoclaved individually ventilated cages (Thoren Caging Systems, Hazelton, PA) with soft-texture bedding (Paperchip, Shepherd Specialty Papers, Watertown, TN) and were provided irradiated chow (Lab Diet 53WU, PMI Nutrition International, Brentwood, MO) and acidified, autoclaved water ad libitum. All cage changes took place in a laminar flow hood. All animal procedures were approved by the University of Missouri IACUC. All animal work was performed in AAALAC -accredited facilities. P. carinii inocula. A cryopreserved P. carinii isolate (no. PRA159, American Type Culture Collection, Manassas, VA) obtained from the lungs of an immuosuppressed Lewis male rat was used for all experiments. To prepare the inoculum, the P. carinii stock was thawed on ice and diluted in DMEM (HyClone, Logan, UT) supplemented with 10% low-endotoxin fetal bovine serum (Cambrex, East Rutherford, NJ) and 10 µg/mL ciprofloxacin. The P. carinii organisms in the inoculum were counted by using a hemocytometer and light microscopy. To prepare the P. carinii filtrate inoculum, half of a prepared P. carinii inoculum was passed successively through 0.8-, 0.45-, and 0.22-µm cellulose acetate membrane syringe filters (Corning Glass Works, Corning, NY). DMEM medium was used to sham-inoculate control rats; each rat received 100 µL of infectious, filtered or sham inoculum. The P. carinii, filtered, and DMEM inocula were tested for the presence of Mycoplasma spp., cilia-associated respiratory bacillus, Strep-

tobacillus moniliformis, Sendai virus, pneumonia virus of mice, sialodacryoadenitis virus (rat coronavirus), mouse hepatitis virus, Kilham rat virus, Toolan H1 virus, rat parvovirus, rat minute virus, minute virus of mice, mouse parvovirus, murine norovirus, mouse thymic virus, reovirus, mouse rotavirus, ectromelia virus, rat theilovirus, Theiler murine encephalomyelitis virus, lymphocytic choriomeningitis virus, lactate dehydrogenase-elevating virus, hantavirus, Seoul virus, rat cytomegalovirus, mouse cytomegalovirus, polyoma virus, K virus, and mouse adenovirus by PCR and for the presence of aerobic and microaerophilic bacteria and fungi by sterility testing (Research Animal Diagnostic Laboratory, Columbia, MO). Necropsy, tissue collection, and histology. At the time points designated in the experimental studies, rats were euthanized by using an inhaled overdose of CO2. Blood was collected by cardiocentesis, and serum was separated and frozen at −20 °C pending serologic evaluation. The abdominal and thoracic cavities were opened aseptically, and tissues were evaluated for gross lesions. The right cranial lung lobe was collected with sterile instruments and frozen at −20 °C pending PCR analysis. For some rat groups, the cut surface of the right cranial lung lobe was cultured prior to freezing. The remaining lung lobes were inflated with neutralbuffered 10% formalin and fixed for at least 24 h, cut transversely into multiple 3- to 4-mm sections, and embedded in paraffin. Embedded lungs were sectioned at 5 µm, placed on glass slides, and stained with hematoxylin and eosin for histologic evaluation or ammoniacal silver for Pneumocystis cysts.7 Histopathologic evaluation of lungs. Idiopathic interstitial pneumonia consistent with that reported for RRV was diagnosed according to established criteria1 by a pathologist viewing lung sections stained with hematoxylin and eosin. Briefly, lung sections that had multiple foci of dense perivascular lymphocytic cuffs in multiple areas of the lung or lymphohistiocytic interstitial pneumonia were classified as positive for idiopathic interstitial pneumonia. Lung sections that had a single focus of dense perivascular lymphocytes that could not be attributed to identifiable conditions were classified as equivocal for idiopathic interstitial pneumonia. Lungs containing other lesions were classified as negative. Silver-stained sections of lungs were evaluated for the presence of 4- to 7-µm, argyrophilic spheres consistent with Pneumocystis cysts.3,8 Pneumocystis PCR and sequencing. PCR for Pneumocystis was performed (Research Animal Diagnostic Laboratory) by using primers designed to conserved regions of the mitochondrial large subunit rRNA gene of P. carinii, P. wakefieldiae, and P. murina. Commercial sequence analysis (SeqWright, Houston, TX) was used to determine the sequence of amplified PCR products, and the resulting sequence was compared with DNA sequences deposited in GenBank by using BLAST software (National Center for Biotechnology Information, Bethesda, MD). Serum antibody testing and lung culture. Sera from rats were tested for antibodies to infectious agents by using MFI2 technology (Research Animal Diagnostic Laboratory). Sera were evaluated for antibodies to the following pathogens: Sendai virus, pneumonia virus of mice, sialodacryoadenitis virus (rat coronavirus), Kilham rat virus, Toolan H1 virus, rat parvovirus, rat minute virus, reovirus, rat theilovirus, lymphocytic choriomeningitis virus, hantavirus, mouse adenovirus, cilia-associated respiratory bacillus, Clostridium piliforme, and Mycoplasma pulmonis.

46

CM10000095.indd 46

3/4/2011 1:34:27 PM

Pneumocystis pneumonia in immunocompetent rats

The right cranial lung lobe was collected from euthanized rats by using sterile technique, the cut surface was inoculated onto 5% sheep blood agar plates, and plates were streaked with a sterile loop. Plates were incubated for 48 h at 35 °C in 7% CO2, after which they were evaluated by a microbiologist for the presence of bacteria. Statistical analysis. Statistical differences were assessed with the Fischer exact test by using SigmaPlot 11.0 software (Systat Software, San Jose, CA). Comparisons with a P value less than or equal to 0.05 were considered statistically significant. Preliminary study: Retrospective analysis for Pneumocystis DNA in paraffin-embedded lung sections from rats with idiopathic interstitial pneumonia consistent with RRV. To explore the possible association between idiopathic lung lesions in rats and P. carinii infection, paraffin-embedded lung tissue from rats with and without idiopathic interstitial pneumonia were tested for the presence of Pneumocystis by PCR. Idiopathic interstitial pneumonia was diagnosed according to established diagnostic criteria1 by a pathologist viewing lung sections stained with hematoxylin and eosin. In addition, silver-stained lungs sections from rats with lesions were evaluated microscopically for the presence of Pneumocystis cysts. Samples collected during 2005 to 2010 from the lungs of 23 rats, 12 to 21 wk old, with idiopathic interstitial pneumonia were evaluated and included rats of at least 6 different strains from 6 different research institutions in the United States. The lungs of 10 Crl:CD(SD) rats (Charles River, Wilmington, MA) and 10 Hsd:SD rats (Harlan Laboratories, Indianapolis, IN), all 14 wk old, that were collected in 2009 and were negative for idiopathic interstitial pneumonia were evaluated also. All lung sections were fixed in neutral-buffered 10% formalin for 24 h before being trimmed and embedded in paraffin. Paraffin-embedded lung sections were stored at room temperature from the time of initial collection until analysis. Sections of paraffin-embedded lung tissue (50 µm) from each rat were made by using a microtome and were placed in sterile microcentrifuge tubes pending DNA extraction and Pneumocystis PCR testing. The microtome blade was cleaned with 70% ethanol between tissue samples from different rats. Experimental infection studies with P. carinii. Experiment 1: Intratracheal inoculation of rats with P. carinii. This experiment was designed to determine whether lesions consistent with idiopathic interstitial pneumonia were induced in immunocompetent rats inoculated intratracheally with P. carinii. Female 4-wk-old Crl:CD(SD) rats (n = 10) were anesthetized with isoflurane and inoculated intratracheally with 100 µL P. carinii inoculum containing 2.1 × 105 organisms; control rats (n = 6) were sham inoculated intratracheally with 100 µL medium. Intratracheal inoculations were performed by suspending the anesthetized rat by the upper incisors in dorsal recumbency on an incline plane, visualizing the laryngeal opening by using a speculum and otoscope.5 The inoculum was instilled in the trachea by using a sterile gel-loading pipette tip and a micropipette, by using materials for intratracheal intubation of rats (Rat Intubation Pack, Hallowell EMC, Pittsfield, MA). All rats were euthanized at 5 wk after inoculation, and the lungs were evaluated by light microscopy for lesions of idiopathic interstitial pneumonia and by PCR for the presence of Pneumocystis DNA. The lungs of rats inoculated with P. carinii were evaluated also for the presence of Pneumocystis cysts by microscopy of silverstained lung sections. Serum was collected from all rats and tested for antibodies to rat viral and bacterial pathogens. The lungs of the P. carinii-inoculated rats were cultured for aerobic bacteria.

Experiment 2: Inoculation of rats with filtered P. carinii inoculum. This experiment was designed to determine whether lesions consistent with idiopathic interstitial pneumonia were induced in immunocompetent rats inoculated intratracheally with a filtered fraction of the P. carinii inoculum. This experiment was conducted at the same time as experiment 1; half of the P. carinii inoculum was unfiltered and used in experiment 1; the remaining half was filtered to exclude all particles whose diameters were greater than 0.22 µm (filtrate) and used in experiment 2. This level of filtration excludes P. carinii and bacteria but not viruses. Female 4-wk-old Crl:CD(SD) rats were anesthetized with isoflurane;10 rats were inoculated intratracheally with 100 µL filtrate, and 6 rats were inoculated intratracheally with 100 µL medium (sham-inoculated controls) as described in experiment 1. All rats were euthanized at 5 wk after inoculation, and the lungs were evaluated by light microscopy for lesions of idiopathic interstitial pneumonia and by PCR for the presence of Pneumocystis DNA. Serum was collected from all rats and tested for antibodies to known viral and bacterial pathogens. Experiment 3: Cohousing of rats with immunosuppressed rats infected with P. carinii. This experiment was designed to determine whether immunocompetent rats acquired P. carinii infection and developed lesions consistent with idiopathic interstitial pneumonia after being cohoused with immunosuppressed rats infected with P. carinii. In addition, the kinetics of P. carinii acquisition and lesion development were assessed. Female 4-wk-old Crl:CD(SD) rats (n = 12) were immunosuppressed by treating with methylprednisolone acetate (Teva, Irvine, CA) subcutaneously once at 40 µg/g body weight during week 1, once at 30 µg/g during week 2, and then weekly at 20 µg/g until week 11. After 3 wk of immunosuppression, rats were anesthetized with isoflurane, and 8 rats were inoculated with 100 µL P. carinii inoculum containing 2.4 × 105 organisms; 4 rats were inoculated intratracheally with 100 µL medium (sham-inoculated controls). At 2 wk after inoculation, P. carinii- and sham-inoculated rats were placed in individual cages, and each was cohoused with two 4-wk-old Crl:CD(SD) rats. At 3, 5, and 7 wk after cohousing, 5 or 6 rats cohoused with P. carinii-inoculated rats and 2 or 3 rats cohoused with sham-inoculated rats were euthanized; the lungs were evaluated by light microscopy for lesions of idiopathic interstitial pneumonia and Pneumocystis cysts and by PCR for the presence of Pneumocystis DNA. Serum was collected from cohoused rats and was tested for antibodies to viral and bacterial pathogens. At the end of the study, the immunosuppressed P. carinii- and sham-inoculated rats were euthanized; the lungs were evaluated for lesions by light microscopy and by PCR for the presence of Pneumocystis DNA.

Results

Retrospective analysis for Pneumocystis DNA in paraffin-embedded lung sections from rats with idiopathic interstitial pneumonia consistent with RRV. To explore the possible association between idiopathic lung lesions in rats and infection with P. carinii, paraffin-embedded lung tissue from rats with and without idiopathic interstitial pneumonia was tested by PCR for the presence of Pneumocystis. The lungs of 12- to 21-wk-old rats (n = 23) with idiopathic interstitial pneumonia were evaluated. These samples were collected in years 2005 to 2010 from rats of at least 6 different strains from 6 different research institutions in the United States. The lungs of 14-wk-old Crl:CD(SD) rats (n = 10) and 14-wk-old Hsd:SD rats (n = 10) that were negative histologi-

47

CM10000095.indd 47

3/4/2011 1:34:27 PM


cally for idiopathic interstitial pneumonia and originated from colonies with a history of being free of idiopathic interstitial pneumonia were evaluated also. Pneumocystis DNA was detected in 20 of the 23 paraffin-embedded lung sections from rats diagnosed with idiopathic interstitial pneumonia and in 0 of the 20 paraffinembedded lung sections from rats negative for idiopathic interstitial pneumonia (P ≤ 0.001; Fisher exact test). The sequence of 11 of the Pneumocystis-positive PCR amplicons was determined, and at least 77 bases of clear sequence were obtained from each amplicon. All 11 sequences were identical to each other over this region, and all were 100% identical to Pneumocystis carinii large subunit rRNA gene sequences deposited in GenBank (EF646865, S79764, U20172, U20171, M58604, S42914, U20170, and U20169). A few Pneumocystis cysts were seen on microscopic evaluation of silver-stained lung sections in 1 of the 23 rats with idiopathic interstitial pneumonia. Inoculation of CD rats with P. carinii. This experiment was conducted to determine whether idiopathic interstitial pneumonia consistent with that of so-called RRV would be induced in immunocompetent CD rats inoculated with P. carinii. Female CD rats (n = 10) were dosed intratracheally with P. carinii and 6 female CD rats were dosed intratracheally with medium (sham-inoculated controls). Rats were euthanized 5 wk after inoculation. Microscopic evaluation of lungs revealed lesions of idiopathic interstitial pneumonia consistent with those reported for RRV infection in 9 of the 10 CD rats inoculated with P. carinii. The remaining P. carinii-inoculated rat had a single dense perivascular lymphocytic cuff and was classified as equivocal for RRV infection. In comparison, none of the 6 sham-inoculated rats was positive for RRV lesions (P ≤ 0.001; Fisher exact test). In the P. carinii-inoculated rats, the histologic severity of lesions varied but consisted of mild to severe multifocal perivascular infiltrates of lymphocytes, plasma cells, and macrophages and mild to severe lymphohistiocytic interstitial pneumonia, with occasional to marked type 2 pneumocyte hyperplasia. The lung lesions induced in the P. carinii-inoculated rats were identical to those seen in spontaneous cases of idiopathic lung lesions that have been attributed to RRV infections (Figure 1 A through D). Pneumocystis DNA was detected in the lungs of 9 of the 10 P. carinii-inoculated rats, whereas all 6 sham-inoculated rats were PCR-negative for Pneumocystis. In addition, Pneumocystis DNA was detected in 8 of the 9 that were positive for RRV lung lesions and in the 1 rat with equivocal lung lesions. Sequence analysis of a subset of the PCR-positive rats (n = 4) showed that the amplicons were 100% identical to P. carinii sequences deposited in GenBank. Histologic evaluation of silver-stained lung sections revealed infrequent Pneumocystis cysts in 4 of the 10 P. carinii-inoculated rats. The cysts were distributed randomly throughout the lung and were not associated with lesions. All Pneumocystis-infected and control rats tested negative for serum antibodies to the infectious viruses and bacteria listed in the Materials and Methods section. The P. carinii and sham inoculums tested negative by PCR for all viruses and bacteria listed in the Materials and Methods section, and no fungi or bacteria were detected by microbiologic culture, except for 18 cfu Enterococcus faecalis bacteria were detected per 100 µL dose of P. carinii inoculum administered to each rat. However, these bacteria were not associated with the development of lung lesions, given that no aerobic bacteria were isolated from the lungs of P. carinii-inoculated rats.

Inoculation of CD rats with filtered P. carinii inoculum. Because P. carinii cannot be propagated efficiently in cell culture, the ATCC isolate used in the current studies was harvested from the lungs of an immunosuppressed rat.23 To rule out the possibility that the P. carinii inoculum contained a virus that was responsible for causing the idiopathic lung lesions in inoculated rats, CD rats (n = 10) were inoculated intratracheally with a 0.22-µm cut-off filtrate of the P. carinii inoculum used in the previous experiment. This level of filtration excludes P. carinii and bacteria but not viruses. Control (sham-inoculated) CD rats (n = 6) were inoculated intratracheally with medium. Rats were euthanized and evaluated 5 wk after inoculation. No lung lesions consistent with those previously reported for RRV infection were seen in either the filtrate- or sham-inoculated rats (Figure 1 E and F). In addition, all filtrate- and sham-inoculated rats tested negative for Pneumocystis by PCR, and all experimental and control rats tested negative for serum antibodies to the infectious viruses and bacteria listed in the Materials and Methods section. Cohousing of CD rats with immunosuppressed rats infected with P. carinii. This experiment was designed to determine whether immunocompetent CD rats acquire P. carinii infection from infected rats and develop lesions consistent with idiopathic interstitial pneumonia. In addition, the kinetics of P. carinii acquisition by naïve rats and lesion development were assessed. Two immunocompetent rats were cohoused with a single immunosuppressed CD rat that had been inoculated with P. carinii or medium (shaminoculated control) 2 wk previously. Confirmation of infection status of the immunosuppressed, dosed rats was confirmed at the end of the cohousing study. All immunosuppressed P. cariniiinoculated rats (n = 8) had lung lesions consistent with Pneumocystis infection and were PCR-positive for Pneumocystis DNA. The immunosuppressed sham-inoculated rats (n = 4) lacked lesions of Pneumocystis infection and were negative for Pneumocystis by PCR. At 3, 5, and 7 wk after cohousing, immunocompetent rats were euthanized and evaluated. At all 3 time points evaluated, lesions of idiopathic interstitial pneumonia consistent with those previously described for RRV infection and Pneumocystis DNA were detected in the lungs of immunocompetent rats cohoused with P. carinii-inoculated rats (Table 1). In contrast, no lesions were seen in rats cohoused with sham-inoculated rats at any time point. Therefore, the lesions described following are most likely the result of infection with P. carinii. Development of lung lesions differed significantly (P ≤ 0.001; Fisher exact test) between rats cohoused with P. carinii-inoculated rats compared with sham-inoculated rats. The number of immunocompetent rats with lungs lesions increased at each time point from 3 to 7 wk, with all rats (n = 6) having characteristic lung disease at 7 wk after cohousing. The severity and character of the lung lesions in immunocompetent rats changed over time (Figure 2 A through F). At 3 wk after cohousing, lesions consisted of mild multifocal perivascular lymphohistiocytic infiltrates with eosinophils, absent to mild perivascular hemorrhage, and mild multifocal lymphohistiocytic interstitial pneumonia. Lesion severity was greatest at 5 wk after cohousing and consisted of moderate to severe multifocal perivascular lymphohistiocytic infiltrates and moderate to severe multifocal lymphohistiocytic interstitial pneumonia. In areas of interstitial pneumonia, there was occasional to marked type 2 pneumocyte hyperplasia, and many alveolar spaces were filled with macrophages. By 7 wk after cohousing,

48

CM10000095.indd 48

3/4/2011 1:34:27 PM


Figure 1. Photomicrographs of lungs from (A, B) immunocompetent F344 rat with naturally occurring idiopathic interstitial pneumonia consistent with so-called RRV and (C, D) immunocompetent CD rat 5 wk after intratracheal inoculation with P. carinii. Note that lesions induced from P. carinii inoculation are indistinguishable from those of naturally acquired lesions previously attributed to RRV infection, presented in panels A and B. (E, F) Lungs from immunocompetent CD rat 5 wk after intratracheal inoculation with P. carinii inoculum passed through a 0.22-µm filter. Note the absence of lesions in these rats inoculated with a preparation that excludes P. carinii but not viruses. Hematoxylin and eosin stain. Bar, 200 µm (A, C, E); 50 µm (B, D, F). Panels B, D, and F are higher-magnification images of panels A, C, and E, respectively.

lung lesions were resolving and consisted of mild multifocal perivascular lymphohistiocytic infiltrates or mild multifocal interstitial pneumonia. Representative images of lungs from immunocompetent CD rats cohoused with sham-inoculated rats at 3, 5, and 7 wk after cohousing are presented (Figure 2 G through L). Histologic evaluation of silver-stained lung sections revealed infrequent Pneumocystis cysts in 5 of the 16 rats cohoused with P. carinii-inoculated rats. The cysts were distributed randomly throughout the lung and were not associated with lesions. Rats cohoused with P. carinii-infected rats tested positive for Pneumocystis DNA at all 3 time points evaluated, with the exception of a single rat that tested negative at the 3 wk time point (Table 1). Two Pneumocystis-positive PCR amplicons were sequenced from each time point. All 6 sequences were 100% identical to Pneumocystis carinii sequences deposited in GenBank. Pneumocystis DNA was detected in all 13 rats with lesions of RRV and in 2 additional rats that were lesion-negative but were cohoused with P. carinii-inoculated rats. All experimental and control cohoused rats tested negative for serum antibodies to the infectious viruses and bacteria listed in the Materials and Methods section.

Discussion

Collectively, the data reported herein prove that P. carinii is an etiologic agent of the idiopathic interstitial pneumonia of rats previously attributed to infection with a putative rat respiratory virus. In retrospective studies, we detected Pneumocystis DNA in 87% of clinical lung samples from rats with histopathologic

evidence of idiopathic interstitial pneumonia and did not detect Pneumocystis DNA in lesion-fee rat lungs. In the subset of PCR amplicons sequenced from these clinical cases, only P. carinii DNA was detected. Evaluation of these naturally occurring cases of RRV showed an association, but not causality, of P. carinii with lung lesions. Cause and effect between P. carinii infection and the development of lung lesions was established in controlled experimental inoculation studies. Lung lesions identical to those previously attributed to infection with RRV were reproduced in rats inoculated intratracheally with a well-characterized P. carinii isolate and those cohoused with P. carinii-inoculated rats. Importantly, there was a strong association between the detection of Pneumocystis DNA, confirmed by sequence analysis to be P. carinii, and the presence of lung lesions. The progression and character of lesions over time paralleled those described for RRV,1 but in our present study, the onset of lesion development was accelerated by at least 2 wk. Our lesions were mild at 3 wk, severe at 5 wk, and resolving by 7 wk after exposure, compared with the previous study1, where the first positive lesions in rats were not seen until 5 wk after exposure but were seen continually until 13 wk. The reason for this difference in lesion onset and progression is unknown but might reflect several factors, including timing of infection, dose of P. carinii, strain of Pneumocystis, strain of rat, and other environmental factors. In our studies, rats were exposed to P. carinii by either direct intratracheal dosing with organisms or indirectly by cohousing with P. carinii-inoculated immunosuppressed rats.

49

CM10000095.indd 49

3/4/2011 1:34:29 PM


Table 1. Detection of lung lesions and Pneumocystis carinii in CD rats cohoused with Pneumocystis carinii-inoculated or sham-inoculated rats Group Cohoused with P. carinii-inoculated rats

Cohoused with sham-inoculated rats

Duration of cohousing (wk)

No. with lung lesions/ total no. of rats

No. PCR-positive for Pneumocystis/ total no. of rats

3

3/5

4/5

5

4/5

5/5

7

6/6

6/6

3

0/2

0/2

5

0/2

0/2

7

0/3

0/3

It is conceivable that in both of these exposure scenarios, rats in our studies received higher and more immediate doses of P. carinii than did the rats in the earlier study,1 which were housed in wire-top cages in the same room with infected immunocompetent rats of variable infection status and exposed periodically to soiled bedding from colony rats. In our studies, a multitude of other potential etiologies were ruled out as being associated with the development of lung lesions. Importantly, a viral etiology for this disease was ruled out when lung lesions were not reproduced in rats dosed with a filtered P. carinii inoculum that allowed viral particles to remain in the inoculum but excluded most other pathogenic microbes including Pneumocystis. In addition, inoculated or cohoused rats that developed lung lesions did not develop serum antibodies to known viral or bacterial pathogens. No known rat or mouse viral, bacterial, or fungal pathogens were detected in the P. carinii inoculum. However, extremely low numbers of E. faecalis bacteria were present in the P. carinii inoculum. Despite this finding, we found no association of E. faecalis or other bacteria with the development of lung lesions, given that no bacteria were cultured from the lungs of immunocompetent rats directly dosed with the P. carinii inoculum. Furthermore, the expected outcome of an inhaled enterococcal bacterial pneumonia would be suppurative bronchopneumonia, which was not seen in inoculated immunocompetent or immunosuppressed rats (data not shown). Although our studies document the role of P. carinii in causing significant lung pathology in immunocompetent rats, whether other Pneumocystis species are capable of causing similar disease is unknown. P. carinii and P. wakefieldiae are the 2 Pneumocystis species that have been documented to infect rats used in research, and at least 3 other provisional Pneumocystis species have been found in wild rats.9,30 P. carinii is the most prevalent Pneumocystis species and can occur as a monoinfection or coinfection with P. wakefieldiae.9,19 Although we found no P. wakefieldiae or other Pneumocystis DNA sequences in the clinical cases we examined, our studies do not rule out the possibility that other Pneumocystis cause pneumonia in immunocompetent rats. Therefore, we recommend excluding all Pneumocystis species from rat colonies in which Pneumocystis pneumonia is unwanted. Experimental studies are ongoing to address the pathogenic potential of P. wakefieldiae in immunocompetent rats. It has long been known that immunocompetent rats can be infected subclinically with Pneumocystis, but the development of Pneumocystis pneumonia was thought to occur only in rats immunodeficient because of genetic (for example, nude rats) or artificial (for example, immunosuppressive doses of glucocorticoids) means.18,19,30 As such, immunocompetent rat colonies have not been tested routinely for Pneumocystis infections. Now that an

etiology for RRV lesions has been identified, Pneumocystis-specific diagnostic tests, such as PCR assays (which already are available), can be used to detect infected rats and monitor colonies for Pneumocystis. Previously, histopathology was the only diagnostic assay available to detect rats with idiopathic pneumonia now known to be Pneumocystis pneumonia.1 High-throughput immunoassays to detect serum antibodies to Pneumocystis would be helpful in testing rat colonies for this pathogen and are in development. An impediment to this process is that a reliable in vitro culture system for propagation of Pneumocystis has not been developed, and in vivo propagation techniques are laborious and require the use of large numbers of immunosuppressed animals. Evaluation of silver-stained lung sections for 4- to 7-µm argyrophilic cysts consistent with those of Pneumocystis is a commonly used histologic technique to diagnose Pneumocystis infections. However, in our studies, cysts were seen only rarely in the lungs of immunocompetent rats with spontaneous Pneumocystis infections, even though 87% of the lungs tested positive for Pneumocystis DNA. In our experimental infection studies, Pneumocystis cysts were seen infrequently and not associated with lesions in about one third of the immunocompetent rats that were positive for lesions and Pneumocystis DNA. Therefore, evaluation of lungs for Pneumocystis cysts in infected immunocompetent rats appears to lack diagnostic sensitivity. The reason for infrequent detection and random distribution of cysts in the lungs of infected immunocompetent rats is unknown, but this finding is consistent with a report of P. murina infection in immunocompetent mice.6 It is an important discovery that Pneumocystis carinii is a cause of the idiopathic lung disease that has plagued immunocompetent rats and confounded research studies for more than a decade. This breakthrough allows the wealth of information regarding the biology, transmission, diagnosis, and control of rodent Pneumocystis infections to be applied to establishing and maintaining rat colonies that are free of Pneumocystis and Pneumocystis pneumonia.8,30,31 Moreover, this finding will allow controlled experiments to be performed to assess how Pneumocystis infections alter physiologic or immunologic responses in immunocompetent rats, possibly confounding research studies beyond inducing severe histologic lung pathology. Our laboratory and others have identified similar idiopathic perivascular and interstitial lung disease in mice (data not shown). We hypothesize that infection with P. murina, the Pneumocystis species specific to mice, is responsible for this similar condition, and investigations are ongoing to test this hypothesis. Previous work suggests this hypothesis is valid, in that mild lung lesions were induced in a low percentage of immunocompetent ICR mice dosed intranasally with lung homogenates containing P. murina.13 In affected mice, mild interstitial monocyte and lympho-

50

CM10000095.indd 50

3/4/2011 1:34:29 PM


Figure 2. (A through F) Photomicrographs of lungs from immunocompetent CD rats cohoused with P. carinii-inoculated immunosuppressed rats at 3, 5 and 7 wk after cohousing. Note the progression of lesions over time. (A, B) 3 wk after cohousing: mild perivascular lymphohistiocytic infiltrates with eosinophils and minimal lymphohistiocytic interstitial pneumonia. (C, D) 5 wk after cohousing: marked multifocal perivascular lymphohistiocytic infiltrates and severe locally extensive lymphohistiocytic interstitial pneumonia. (E, F) 7 wk after cohousing: mild perivascular lymphohistiocytic infiltrates and minimal lymphohistiocytic interstitial pneumonia. (G through L) Photomicrographs of lungs from immunocompetent CD rats cohoused with sham-inoculated rats at 3, 5 and 7 wk after cohousing. Note the absence of lesions at all time points. Hematoxylin and eosin stain. Bar, 200 µm (A, C, E, G, I, K); 50 µm (B, D, F, H, J, L). Panels B, D, F, H, J, and L are higher-magnification images of panels A, C, E, G, I, and K, respectively.

51

CM10000095.indd 51

3/4/2011 1:34:32 PM


cyte infiltrates were seen 1 and 2 wk after inoculation, and only scattered mononuclear cell infiltrates were seen in alveolar septa at 3 and 4 wk after inoculation. In another study, when P. murina was transmitted from infected immunocompetent BALB/c mice to naïve cohoused BALB/c mice, moderate infiltrates of mononuclear cells were seen around bronchi or vessels of the lungs of newly infected mice.6 The marked, predictable lung pathology in immunocompetent rats experimentally infected with P. carinii provides a possible animal model of P. jirovecii infections in immunocompetent humans. It is well known that P. jirovecii infections cause severe pneumonia in immunosuppressed humans and are a leading cause of death in patients infected with HIV. However, evidence is starting to be amassed associating subclinical P. jirovecii infections in immunocompetent humans with several diseases in infants and adults, including sudden infant death syndrome, chronic obstructive pulmonary disease, asthma, bronchiolitis, and other lung disease states.20,21 Pneumocystis-infected immunocompetent rats provide a potential rodent model in which to investigate Pneumocystis infections in immunocompetent hosts. In summary, our data show that Pneumocystis, and not a virus, is responsible for causing lung lesions in rats that were previously classified as idiopathic. This finding makes RRV an incorrect moniker in reference to this disease. We suggest that this disease be called Pneumocystis pneumonia, thereby following the convention of the Pneumocystis literature.23 Therefore, the disease is Pneumocystis pneumonia, not RRV pneumonia.

Acknowledgments

We thank Greg Purdy, Giedre Turner, Maria Evola, Michael Drake, and Marissa Wolfe for technical assistance and Howard Wilson for photography.

References

1. Albers TM, Simon MA, Clifford CB. 2009. Histopathology of naturally transmitted ‘rat respiratory virus’: progression of lesions and proposed diagnostic criteria. Vet Pathol 46:992–999. 2. Aliouat EM, Mazars E, Dei-Cas E, Delcourt P, Billaut P, Camus D. 1994. Pneumocystis cross-infection experiments using SCID mice and nude rats as recipient hosts showed strong host–species specificity. J Eukaryot Microbiol 41:71S. 3. Aliouat-Denis CM, Martinez A, Aliouat el M, Pottier M, Gantois N, Dei-Cas E. 2009. The Pneumocystis life cycle. Mem Inst Oswaldo Cruz 104:419–426. 4. Besselsen DG, Franklin CL, Livingston RS, Riley LK. 2008. Lurking in the shadows: emerging rodent infectious diseases. ILAR J 49:277–290. 5. Boylan CJ, Current WL. 1992. Improved rat model of Pneumocystis carinii pneumonia: induced laboratory infections in Pneumocystis-free animals. Infect Immun 60:1589–1597. 6. Chabe M, Dei-Cas E, Creusy C, Fleurisse L, Respaldiza N, Camus D, Durand-Joly I. 2004. Immunocompetent hosts as a reservoir of Pneumocystis organisms: histological and RT-PCR data demonstrate active replication. Eur J Clin Microbiol Infect Dis 23:89–97. 7. Churukian CJ, Schenk EA. 1988. Staining Pneumocystis carinii and fungi in unfixed specimens with ammoniacal silver using a microwave oven. J Histotechnol 11:19–21. 8. Cushion MT. 2007. Pneumocystis, p 1789–1801. In: Murray PR, Baron Ej, Jorgensen JH, Landry ML, Pfaller MA, editors. Manual of clinical microbiology. Washington (DC): ASM Press.

9. Cushion MT, Keely SP, Stringer JR. 2004. Molecular and phenotypic description of Pneumocystis wakefieldiae sp. nov., a new species in rats. Mycologia 96:429–438. 10. Durand-Joly I,. Aliouat el M, Recourt C, Guyot K, Francois N, Wauquier M, Camus D, Dei-Cas E. 2002. Pneumocystis carinii f. sp. hominis is not infectious for SCID mice. J Clin Microbiol 40:1862–1865. 11. Elwell MR, Mahler JF, Rao GN. 1997. Have you seen this? Inflammatory lesions in the lungs of rats. Toxicol Pathol 25:529–531. 12. Farrar P, LaRegina M. 1997. Diagnostic exercise: interstitial pneumonia in viral and mycoplasmal antibody-free Sprague Dawley rats. ACLAD News 18:5–9. 13. Furuta T, Ueda K, Fujiwara K, Yamanouchi K. 1985. Cellular and humoral immune responses of mice subclinically infected with Pneumocystis carinii. Infect Immun 47:544–548. 14. Gigliotti F, Haidaris PJ, Haidaris CG, Wright TW, Van der Meid KR. 1993. Further evidence of host species-specific variation in antigens of Pneumocystis carinii using the polymerase chain reaction. J Infect Dis 168:191–194. 15. Gigliotti F, Harmsen AG, Haidaris CG, Haidaris PJ. 1993. Pneumocystis carinii is not universally transmissible between mammalian species. Infect Immun 61:2886–2890. 16. Gilbert BE, Black MB, Waldrep JC, Bennick JB, Montgomery C, Knight V. 1997. Cyclosporin A liposome aerosol: lack of acute toxicity in rats with a high incidence of underlying pneumonitis. Inhal Toxicol 9:717–730. 17. Hughes WT. 1982. Natural mode of acquisition for de novo infection with Pneumocystis carinii. J Infect Dis 145:842–848. 18. Icenhour CR, Rebholz SL, Collins MS, Cushion MT. 2001. Early acquisition of Pneumocystis carinii in neonatal rats using targeted PCR and oral swabs. J Eukaryot Microbiol Suppl:135S–136S. 19. Icenhour CR, Rebholz SL, Collins MS, Cushion MT. 2001. Widespread occurrence of Pneumocystis carinii in commercial rat colonies detected using targeted PCR and oral swabs. J Clin Microbiol 39:3437–3441. 20. Morris A, Wei K, Afshar K, Huang L. 2008. Epidemiology and clinical significance of Pneumocystis colonization. J Infect Dis 197:10–17. 21. Peterson JC, Cushion MT. 2005. Pneumocystis: not just pneumonia. Curr Opin Microbiol 8:393–398. 22. Pritchett-Corning KR, Cosentino J, Clifford CB. 2009. Contemporary prevalence of infectious agents in laboratory mice and rats. Lab Anim 43:165–173. 23. Redhead SA, Cushion MT, Frenkel JK, Stringer JR. 2006. Pneumocystis and Trypanosoma cruzi: nomenclature and typifications. J Eukaryot Microbiol 53:2–11. 24. Riley LK, Purdy G, Dodds J, Franklin C, Besch-Williford C, Hook RR, Wagner JE. 1997. Idiopathic lung lesions in rats: search for an etiologic agent. Contemp Top Lab Anim Sci 36:46. 25. Riley LK, Simmons JH, Purdy G, Livingston R, Franklin C, BeschWilliford C. 1999. Research update: idiopathic lung lesions in rats. ACLAD Newsl 20:9–11. 26. Slaoui M, Dreef HC, van Esch E. 1998. Inflammatory lesions in the lungs of Wistar rats. Toxicol Pathol 26:712–713, discussion 714. 27. Walzer PD, Schnelle V, Armstrong D, Rosen PP. 1977. Nude mouse: a new experimental model for Pneumocystis carinii infection. Science 197:177–179. 28. Walzer PD, Schultz MG, Western KA, Robbins JB. 1973. Pneumocystis carinii pneumonia and primary immune deficiency diseases of infancy and childhood. J Pediatr 82:416–422. 29. Walzer PD, Schultz MG, Western KA, Robbins JF. 1976. Pneumocystis carinii pneumonia and primary immune deficiency diseases. Natl Cancer Inst Monogr 43:65–74. 30. Weisbroth SH. 2006. Pneumocystis: newer knowledge about the biology of this group of organisms in laboratory rats and mice. Lab Anim (NY) 35:55–61. 31. Weisbroth SH, Kohn DF, Boot R. Bacterial, mycoplasmal, and mycotic infections. In: Suckow MA, Weisbroth SH, Franklin CL, editors. The laboratory rat. New York (NY): Academic Press.

52

CM10000095.indd 52

3/4/2011 1:34:33 PM