JOURNAL OF CLINICAL MICROBIOLOGY,0095-1137/99/$04.0010

Mar. 1999, p. 620–627 Vol. 37, No. 3

Copyright © 1999, American Society for Microbiology. All Rights Reserved.

Mycoplasma hyopneumoniae Potentiation of PorcineReproductive and Respiratory Syndrome

Virus-Induced PneumoniaEILEEN L. THACKER,1* PATRICK G. HALBUR,2 RICHARD F. ROSS,1

ROONGROJE THANAWONGNUWECH,2,3 AND BRAD J. THACKER2

Veterinary Medical Research Institute1 and Veterinary Diagnostic and Production Animal Medicine,2 College ofVeterinary Medicine, Iowa State University, Ames, Iowa 50011, and Department of Veterinary Pathology,

Faculty of Veterinary Science, Chulalongkorn University, Bangkok, 10330, Thailand3

Received 12 August 1998/Returned for modification 19 October 1998/Accepted 11 December 1998

An experimental model that demonstrates a mycoplasma species acting to potentiate a viral pneumonia wasdeveloped. Mycoplasma hyopneumoniae, which produces a chronic, lymphohistiocytic bronchopneumonia inpigs, was found to potentiate the severity and the duration of a virus-induced pneumonia in pigs. Pigs wereinoculated with M. hyopneumoniae 21 days prior to, simultaneously with, or 10 days after inoculation withporcine reproductive and respiratory syndrome virus (PRRSV), which induces an acute interstitial pneumoniain pigs. PRRSV-induced clinical respiratory disease and macroscopic and microscopic pneumonic lesions weremore severe and persistent in M. hyopneumoniae-infected pigs. At 28 or 38 days after PRRSV inoculation, M.hyopneumoniae-infected pigs still exhibited lesions typical of PRRSV-induced pneumonia, whereas the lungs ofpigs which had received only PRRSV were essentially normal. On the basis of macroscopic lung lesions, itappears that PRRSV infection did not influence the severity of M. hyopneumoniae infection, although micro-scopic lesions typical of M. hyopneumoniae were more severe in PRRSV-infected pigs. These results indicatethat M. hyopneumoniae infection potentiates PRRSV-induced disease and lesions. Most importantly, M. hyo-pneumoniae-infected pigs with minimal to nondetectable mycoplasmal pneumonia lesions manifested signifi-cantly increased PRRSV-induced pneumonia lesions compared to pigs infected with PRRSV only. This dis-covery is important with respect to the control of respiratory disease in pigs and has implications in elucidatingthe potential contribution of mycoplasmas in the pathogenesis of viral infections of other species, includinghumans.

Traditionally it has been presumed that viral infections po-tentiate and increase the susceptibility of a host to bacterialinfections. However, recent evidence indicates that membersof the class of Mollicutes, which includes the mycoplasmas andureaplasmas, may play a more significant primary role in viraldiseases than was previously realized. In 1989, it was reportedthat a high proportion of AIDS patients were infected withMycoplasma fermentans and Mycoplasma penetrans (4, 16).Since then, several other mycoplasmas, including Mycoplasmagenitalium and Mycoplasma pirum, have been implicated inplaying a synergistic role with human immunodeficiency virus(HIV) (2, 4). Evidence that supports the role that mycoplasmasplay in accelerating the progression of HIV to AIDS has in-cluded the enhanced cytopathic effects (CPE) on tracheal ep-ithelial cells (25) and human CD41 lymphocytes (2) infectedwith both HIV and M. fermentans in vitro. The possible mech-anisms by which mycoplasmas could influence the pathogene-sis of HIV have not yet been elucidated, although mycoplas-mas have been shown to activate B and T cells polyclonallyboth in vitro and in vivo (20, 24).

Recently, a new respiratory syndrome in swine, designatedporcine respiratory disease complex (PRDC), has emerged asa serious health problem in most pig-raising regions of theworld. Pneumonia in pigs with PRDC is due to a combinationof both viral and bacterial agents. Mycoplasma hyopneumoniae

and porcine reproductive and respiratory syndrome virus(PRRSV), an Arterivirus (5), are two of the most commonpathogens isolated from pigs exhibiting PRDC.

M. hyopneumoniae is recognized as the causative agent ofporcine enzootic pneumonia, a mild, chronic pneumonia com-monly complicated by opportunistic infections with other bac-teria (22). The primary clinical sign associated with M. hyo-pneumoniae infection is a sporadic, dry, nonproductive cough.Other clinical signs, such as fever or impaired growth, arelinked to secondary invaders, especially Pasteurella multocida.Typical mycoplasmal pneumonia lesions consist of well-demar-cated dark-red-to-purple (acute) or tan-grey (chronic) areas ofcranioventral consolidation. Microscopic examination revealsbronchopneumonia with suppurative and histiocytic alveolitiswith peribronchiolar and perivascular lymphohistiocytic cuffingand nodule formation, typical of hyperplasia of bronchoalveo-lar lymphoid tissue.

In contrast, PRRSV induces a severe, acute pneumonia withclinical disease characterized by labored and accentuated ab-dominal respiration and tachypnea. Coughing is rarely ob-served. In addition, pigs infected with PRRSV exhibit elevatedrectal temperatures, with pronounced lethargy and anorexia.Gross lesions associated with PRRSV infection consist of se-vere, multifocal to diffuse, tan-mottled consolidation of thelung. Microscopic lesions include septal infiltration with mono-nuclear cells, type II pneumocyte hypertrophy and hyperplasia,and an alveolar exudate consisting of mixed inflammatory cellsand necrotic macrophages.

The mechanisms by which PRRSV and M. hyopneumoniaecause disease are quite different. PRRSV infects cells of the

* Corresponding author. Mailing address: VMRI, Iowa State Uni-versity, 1802 Elwood Dr., Ames, IA 50011. Phone: (515) 294-5097. Fax:(515) 294-1401. E-mail: ethacker@iastate.edu.

620

on May 2, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

macrophage/monocyte/dendritic lineage. Pulmonary alveolarmacrophages (PAMs) and pulmonary intravascular macro-phages (PIMs) are the primary sites of replication in the lung(12, 28). Infection of PAMs and PIMs by PRRSV induces celllysis, presumably resulting in a decreased ability of the respi-ratory tract to defend against both respiratory and systemicpathogens (9). In contrast, M. hyopneumoniae attaches to thecilia of tracheal epithelial cells, resulting in damage to epithe-lial cells and the mucociliary apparatus (6). The mechanismsexplaining the lymphocytic infiltration observed with M. hyo-pneumoniae are unknown. The resulting lesions lead to con-solidation of lung parenchyma and local invasion of opportu-nistic bacteria or viruses that can ultimately lead to systemicdisease.

Consistent with other virus-bacterium interaction models,the most commonly proposed hypothesis is that PRRSV is theprimary pathogen and bacteria are secondarily involved in thepathogenesis of PRDC. In the study reported here, we foundthat infection with M. hyopneumoniae potentiated and pro-longed PRRSV-induced pneumonia clinically, macroscopicallyand microscopically. This report describes an in vivo modelthat conclusively demonstrates a potentiating effect on a viralinfection by a mycoplasma species.

MATERIALS AND METHODS

Pigs. One hundred forty PRRSV- and mycoplasma-free crossbred (Landrace,large white, and duroc) pigs were obtained from a commercial herd at the agesof 10 to 14 days and were randomly assigned to seven treatment groups. Thestudy was conducted in accordance with the guidelines of the Iowa State Uni-versity Institutional Committee on Animal Care and Use.

Inocula and experimental design. The experimental design is summarized inTable 1. The pigs were 6 weeks old at day 0 (the day on which three of the fourPRRSV groups were inoculated with PRRSV). An inoculating dose of 105 50%tissue culture infective doses (TCID50) of the high-virulence PRRSV strainATCC VR-2385, passage 6, in a 5-ml volume was administered intranasally topigs in groups A, B, C, and F (12). A tissue homogenate containing a derivativeof M. hyopneumoniae 11 (105 color-changing units [CCU] per ml) was adminis-tered intratracheally to pigs in groups A, C, and E at a dilution of 1:100 in 10 mlof mycoplasmal Friis medium (23). M. hyopneumoniae was administered at adilution of 1:50 in 5 ml of medium to pigs in groups B and D due to the youngerage (3 weeks) and smaller size of the pigs at the time of inoculation.

Serology. Blood was collected periodically throughout the trial in order toevaluate antibody production. All sera were tested for antibodies against PRRSVwith a commercially available enzyme-linked immunofluorescent assay (ELISA)(HerdChek: PRRS; IDEXX Laboratories, Inc., Westbrook, Maine) according tothe procedures described by the manufacturer. Samples were considered positiveif the calculated sample-to-positive (S/P) ratio was 0.4 or greater. M. hyopneu-moniae antibody titers were determined by ELISA as previously described (3).Known positive and negative sera were included as controls in each plate.Readings more than 2 standard deviations above the mean value of the negativecontrol were considered positive.

PRRSV and M. hyopneumoniae isolation and titration. PRRSV isolation wasperformed by using bronchoalveolar lavage (BAL) fluid obtained at necropsy bylavaging the bronchi with 50 ml of minimal essential medium (MEM) containingantibiotics (9 mg of gentamicin/ml, 100 U of penicillin G/ml, and 100 mg ofstreptomycin/ml) (18). Virus isolation was then performed according to an es-tablished protocol (17). Virus titration was performed on 10% (wt/vol) homog-

enized lung tissues in MEM as previously described with several modifications(8). Briefly, 200 ml of 10-fold serial dilutions of lung homogenate was inoculatedonto a confluent monolayer of CRL11171 cells for 1 h at 37°C in a humidifiedatmosphere with 5% CO2. The cultures were monitored daily for CPE. If CPEwas not observed within 7 days, the cultures were frozen and thawed and blindlypassaged three times to be considered negative. Monolayers were stained with ananti-PRRSV monoclonal antibody, SDOW-17 (South Dakota State University,Brookings), followed by fluorescein isothiocyanate-conjugated anti-mouse im-munoglobulin and were viewed with a fluorescence microscope for evidence ofspecific viral antigens (14). M. hyopneumoniae was isolated from lung sectionsand titrated as previously described (22). Mycoplasma-appearing colonies thatdeveloped were specifically identified by using epiimmunofluorescence with con-jugates prepared from rabbit antisera to M. hyopneumoniae 11 (7).

Clinical evaluation. Pigs were evaluated daily for at least 15 min for clinicalsigns, including appetite, cough, increased respiration rate, or behavioralchanges. A daily clinical respiratory score was assessed on days 0 to 14 accordingto a previously described system (14).

Pathologic examination. Pigs were necropsied at either day 3, day 10, or day28, as outlined in Table 1. The right rib cage was reflected, and the lungs wereremoved and evaluated for macroscopic lesions. A portion of lung was asepticallycollected for M. hyopneumoniae and PRRSV isolation, fluorescent antibodyassay (FA), immunohistochemistry (IHC), and histopathologic examination. Thelungs were then lavaged, and BAL fluid was obtained. Lesions consistent withmycoplasmal pneumonia (dark-red-to-purple consolidated areas) were sketchedon a standard lung diagram. The proportion of lung surface with lesions wasdetermined from the diagram by using a Zeiss SEM-IPS image analyzing system(23). In contrast to mycoplasma-induced lesions, PRRSV-infected lungs werecharacterized by parenchyma that was mottled tan and rubbery and failed tocollapse. The lung lesions were scored by using a previously developed systembased on the approximate volume that each lobe contributes to the entire lung:the right cranial lobe, right middle lobe, cranial part of the left cranial lobe, andcaudal part of the left cranial lobe each contribute 10% each of the total lungvolume, the accessory lobe contributes 5%, and the right and left caudal lobeseach contribute 27.5% (13). These scores were then used to calculate the totallung lesion score based on the relative contribution of each lobe.

Sections were taken from all lung lobes, fixed in 10% neutral buffered forma-lin, and routinely processed and embedded in paraffin in an automated tissueprocessor. Lung sections were blindly examined and given a score (0 to 4) forperibronchiolar and perivascular lymphoid cuffing and nodule formation consis-tent with M. hyopneumoniae-induced pneumonia lesions. The severity ofPRRSV-induced and interstitial pneumonia lesions was also scored (0 to 6). Thescoring systems are summarized in Table 4, footnotes a and b.

PRRSV and M. hyopneumoniae antigen detection. PRRSV-specific antigen wasdetected in lung tissues by a previously described IHC method (10). IHC wasperformed on sections cut from one paraffin-embedded lung tissue block whichincluded three pieces (1 by 2 cm) of lung, one each from the left cranial,accessory, and caudal lobes. The number of PRRSV antigen-positive cells wascounted as described elsewhere. A direct immunofluorescence procedure wasused for detection of M. hyopneumoniae as described previously (1).

Statistics. Data were subjected to analysis of variance (ANOVA). If the Pvalue from the ANOVA was less than or equal to 0.05, pairwise comparisons ofthe different treatment groups were performed by least significant difference atthe P , 0.05 rejection level.

RESULTS

Clinical disease. Clinical respiratory disease data are pre-sented in Table 2. All groups inoculated with PRRSV dis-played signs of respiratory disease consistent with PRRSV-induced pneumonia by day 3 postinoculation. Signs includedlabored breathing and increased respiratory rate. Coughingwas observed in all M. hyopneumoniae-inoculated groups be-

TABLE 1. Experimental design—infection status, sequence of inoculation, and number of pigs necropsied on each of three days

Group Infection status (day of inoculationa)No. of pigs necropsied at: Total no. of pigs

necropsiedDay 3 Day 10 Day 28

A PRRSV (0), M. hyopneumoniae (0) 6 6 8 20B PRRSV (0), M. hyopneumoniae (221) 6 6 8 20C PRRSV (210), M. hyopneumoniae (0) 6 6 8 20D M. hyopneumoniae (221) 6 6 8 20E M. hyopneumoniae (0) 6 6 8 20F PRRSV (0) 6 6 8 20G Control (0) 6 6 8 20

a Pigs were 6 weeks old at day 0.

VOL. 37, 1999 M. HYOPNEUMONIAE AND PRRSV-INDUCED PNEUMONIA 621

on May 2, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

ginning at days 10 to 14 but not in pigs infected with PRRSVonly (group F).

Analysis of the respiratory disease scores indicated that pigsinfected with both M. hyopneumoniae and PRRSV had moresevere clinical respiratory disease than the single-organism-infected groups. Group B, which was inoculated with M. hyo-pneumoniae 21 days prior to PRRSV inoculation, developedmore severe respiratory disease within the first 3 days afterPRRSV inoculation than any other PRRSV-infected groups.Overall, the pigs in groups A and B, which had received bothPRRSV and M. hyopneumoniae, had significantly more severeclinical respiratory disease than pigs inoculated with PRRSValone (group F) or M. hyopneumoniae alone (groups D and E).

Respiratory scores from pigs in group C do not match thoseof the other PRRSV-infected groups due to the 10-day intervalbetween the inoculation of group C pigs with PRRSV and thefirst determination of clinical scores shown in Table 2. Prior toinoculation with M. hyopneumoniae on day 0, the clinical scoresof group C matched those of group F for the first 10 daysfollowing PRRSV infection, which for group C would be days210 to 0 (data not shown). The respiratory scores for group Cin Table 2 were determined 10 additional days post-PRRSVchallenge. Thus, the score of 2.2 6 0.8 for group C on days 1to 3 in Table 2 was actually determined 11 to 13 days post-PRRSV challenge and could be compared to the scores in the

final column, determined on days 11 to 14, for the otherPRRSV-challenged groups. If the group C data on days 1 to 3are compared to the data for the other PRRSV-challengedgroups on days 11 to 14, the clinical disease in group C isequivalent to that in group F. However, even though there isno comparable PRRSV-only control that matches group C formore than 14 days postinoculation, the clinical respiratory dis-ease in group C shows no evidence of regressing over thecourse of the monitoring period, extending to 24 days afterinoculation with PRRSV.

Macroscopic lesions. The mean percentages of lung tissuewith visible pneumonia, either PRRSV-induced or M. hyo-pneumoniae-induced pneumonia, are summarized in Table 3.At necropsy on day 3, PRRSV-infected pigs were already ex-hibiting PRRSV-induced pneumonia. PRRSV-induced lesionsconsisted of lungs which were mottled tan or diffusely tan, werefirmer and heavier than normal lungs, and failed to collapseupon removal from the chest cavity (Fig. 1b), compared to anormal lung (Fig. 1a). The pneumonia was well developed ingroup C, which had been infected with PRRSV 13 days prior tonecropsy. No visible M. hyopneumoniae-induced pneumoniawas observed at day 3.

At day 10, the level of PRRSV-induced pneumonia weresimilar in groups A, B, and F, all of which had receivedPRRSV on day 0. The percentage of pneumonic lung lesionswas lower in group C at day 10. However, pigs in group C,inoculated with PRRSV 10 days prior to receiving M. hyopneu-moniae on day 0, were 20 days post-PRRSV challenge. Inter-estingly, groups A and C had higher percentages of pneumoniclung lesions due to M. hyopneumoniae than the M. hyopneu-moniae-only group (group E). In addition, the mean percent-ages of lung tissue with mycoplasmal pneumonia in groups Aand C were greater than those in either group B or group D,which had been inoculated with M. hyopneumoniae 31 daysprior to necropsy. M. hyopneumoniae-induced lesions were eas-ily differentiated from the PRRSV-induced lesions, as theyconsisted of well-demarcated areas of dark-red to purple firmparenchyma, located primarily at the ventral and cranial tips ofthe anterior and middle lobes (Fig. 1c).

At day 28, PRRSV-induced pneumonia was observed in onlytwo of the eight pigs in the group inoculated with PRRSV only(group F). In contrast, PRRSV-induced lung lesions were ob-served in all pigs in the dually infected groups (groups A, B,and C) (Fig. 1d). The percentage of lung tissue with PRRSV-induced pneumonia lesions in group A was significantly greaterthan that in group B or C. The levels of mycoplasmal pneu-

TABLE 2. Summary of average clinical respiratory disease scores ofpigs by group after inoculation with PRRSV, M. hyopneumoniae,

or both

Group

Scorea at:

Day 1–3b

(n 5 20)Day 4–10(n 5 14)

Day 11–14(n 5 8)

A 0.8 6 0.4 B 2.4 6 0.7 A 4.1 6 0.5 AB 1.6 6 0.4 A 2.4 6 0.5 A 3.7 6 0.6 AC 2.2 6 0.8 A 2.6 6 0.5 A 2.9 6 0.6 BD 0 D 0 C 0 CE 0 D 0 C 0 CF 0.6 6 0.3 C 1.9 6 0.4 B 2.1 6 1.4 BG 0 D 0 C 0 C

a Scores: 0, normal; 1, mild dyspnea and/or tachypnea when stressed; 2, milddyspnea and/or tachypnea when at rest; 3, moderate dyspnea and/or tachypneawhen stressed; 4, moderate dyspnea and/or tachypnea when at rest; 5, severedyspnea and/or tachypnea when stressed; 6, severe dyspnea and/or tachypneawhen at rest. Within each column, values followed by different letters (A, B, C,or D) are significantly different from each other (P , 0.001).

b Pigs were 6 weeks old at day 0.

TABLE 3. Percentage of lung with visible pneumonia lesions in pigs infected with either M. hyopneumoniae, PRRSV, or both

Group

% of lung exhibiting pneumonia induced by PRRSVa or M. hyopneumoniaeb at the following necropsy date:

Day 3 Day 10 Day 28

PRRSV M. hyopneumoniae PRRSV M. hyopneumoniae PRRSV M. hyopneumoniae

A 8.2 6 4.5 Bc 0.2 6 0.3 50.8 6 15.9 A 4.1 6 4.1 A 42.9 6 7.5 A 7.7 6 5.7 AB 12.3 6 6.0 B 1.6 6 1.7 48.8 6 26.4 A 1.3 6 2.2 B 15.6 6 14.5 C 3.3 6 5.9 B,CCd 42.8 6 12.8 A 0.01 6 0.02 29.5 6 9.1 B 2.5 6 3.1 A,B 28.8 6 14.1 B 7.2 6 8.1 A,BD 0 C 0.1 6 0.1 0 C 0.9 6 2.1 B 0 D 1.2 6 3.0 C,DE 0 C 0.2 6 0.3 0 C 0.2 6 0.2 B 0 D 8.3 6 4.0 AF 13.7 6 5.8 B 0.1 6 0.2 56.5 6 12.7 A 0.05 6 0.1 B 0.8 6 1.59 D 0 DG 0 C 0.1 6 0.3 0 C 0.02 6 0.05 B 0 D 0 D

a As estimated by visual observation.b As determined by lesion sketches and image analysis.c Within each column, values followed by different letters (A, B, C, or D) are significantly different from each other (P , 0.001).d Group C received PRRSV on day 210 and M. hyopneumoniae on day 0 and so is not matched with the other groups with respect to the number of days after

inoculation with PRRSV.

622 THACKER ET AL. J. CLIN. MICROBIOL.

on May 2, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

monia were similar in groups A, C, and E, while groups B andD had minimal mycoplasmal lesions.

Microscopic data. Histopathological evaluations are pre-sented in Table 4. Interstitial pneumonia consistent with

PRRSV-induced pneumonia was present in all PRRSV-in-fected pigs at day 3. This pneumonia was characterized by type2 pneumocyte hypertrophy and hyperplasia, septal infiltrationwith monocytes, and increased alveolar exudate consisting of

FIG. 1. (a) Normal lungs from an uninfected pig. (b) Lungs from a pig infected 10 days previously with PRRSV. The lungs fail to collapse and are diffusely mottledand tan in appearance. (c) Lungs from a pig infected 28 days previously with M. hyopneumoniae. Lungs have multiple well-demarcated, dark-red areas of pneumoniain the cranioventral region. (d) Lungs from a pig dually infected 28 days previously with PRRSV and M. hyopneumoniae, exhibiting both the characteristic failure tocollapse and mottled tan appearance of PRRSV and the well-demarcated, dark-red consolidated areas typical of M. hyopneumoniae.

TABLE 4. Microscopic lesion scores from pigs inoculated with either PRRSV, M. hyopneumoniae, or both

Group

Score for pneumonia induced by PRRSVa or M. hyopneumoniaeb at the following necropsy date:

Day 3 Day 10 Day 28

PRRSV M. hyopneumoniae PRRSV M. hyopneumoniae PRRSV M. hyopneumoniae

A 1.2 6 1.1 Bc 0 4.0 6 0.9 A 1.7 6 1.0 A 4.8 6 0.7 A 3.9 6 0.4 AB 1.0 6 0 B 0.3 6 0.5 3.3 6 1.5 A 0.8 6 0.4 B 1.3 6 0.5 B 1.6 6 1.1 CCd 3.3 6 0.5 A 0 3.3 6 0.5 A 1.8 6 1.0 A 4.3 6 1.2 A 3.8 6 0.5 AD 0 C 0.2 6 0.4 0 B 0.5 6 0.6 B,C 0.1 6 0.4 C 1.0 6 0 CE 0.2 6 0.4 C 0 0 B 0.5 6 0.6 B,C 0 C 2.5 6 0.8 BF 1.5 6 0.8 B 0.3 6 0.5 3.7 6 0.8 A 0 C 1.4 6 0.5 B 1.1 6 0.6 CG 0.2 6 0.4 C 0 0 B 0.8 6 0.4 B,C 0 C 0.3 6 0.5 D

a PRRSV scores are based on the severity of interstitial pneumonia. 0, no microscopic lesions; 1, mild multifocal pneumonia; 2, mild diffuse pneumonia; 3, moderatemultifocal pneumonia; 4, moderate diffuse pneumonia; 5, severe multifocal pneumonia; 6, severe diffuse pneumonia.

b M. hyopneumoniae scores are based on the severity of peribronchiolar and perivascular cuffing and lymphoid nodule formation, as follows: 1, mild; 2, moderate;3, severe; 4, very severe. 0, no microscopic lesions.

c Within each column, values followed by different letters (A, B, C, or D) are significantly different from each other (P , 0.001).d Group C received PRRSV on day 210 and M. hyopneumoniae on day 0 and so is not matched with the other groups with respect to the number of days after

inoculation with PRRSV.

VOL. 37, 1999 M. HYOPNEUMONIAE AND PRRSV-INDUCED PNEUMONIA 623

on May 2, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

macrophages, necrotic macrophages, multinucleated cells, andproteinaceous fluid. Groups A and B had microscopic lesionsequivalent to those observed in pigs from group F (PRRSVonly), whereas group C had more severe PRRSV-induced le-sions, as expected due to its longer interval following PRRSVinoculation (13 days). M. hyopneumoniae-induced lung lesionswere not yet observed at day 3.

Microscopic lesions consistent with PRRSV-induced inter-stitial pneumonia were similar in all PRRSV-infected pigs onday 10 (Fig. 2b). Peribronchiolar and perivascular lymphohis-tiocytic cuffing consistent with M. hyopneumoniae-inducedpneumonia was observed and was significantly more severe ingroups A and C than in group E at day 10 (Fig. 2c).

Groups A and C had significantly more severe PRRSV-induced interstitial pneumonia than group F (PRRSV only) atday 28. Microscopic lesions consistent with M. hyopneumoniaeinfection were also more severe in the dually infected groups Aand C than in group E (M. hyopneumoniae only) at day 28 (Fig.2d).

The mild peribronchiolar and perivascular lymphohistiocyticcuffing reported in the negative-control group and group F(PRRSV only) are nonspecific changes. There was no evidenceof M. hyopneumoniae infection in either the negative-controlgroup or group F based on culture, which is a more sensitive

test for the presence of M. hyopneumoniae. Peribronchiolarlymphoid hyperplasia is considered the most characteristic le-sion associated with M. hyopneumoniae infection; however, incases of mycoplasmal pneumonia, it typically progresses todiscrete peribronchiolar lymphoid nodule or follicle formation.However, these changes are not found only in pigs infectedwith mycoplasmas. Peribronchiolar lymphohistiocytic cuffinghas been reported in pigs experimentally infected with PRRSV(13) or may simply be associated with chronic antigen stimu-lation from the environment. PRRSV-induced peribronchiolarlymphohistiocytic cuffing does not progress to discrete noduleformation and typically resolves by day 28 postinfection.

PRRSV and M. hyopneumoniae isolation and titration.PRRSV was isolated from BAL of all groups of PRRSV-infected pigs. PRRSV was isolated from the greatest numberof pigs in groups A (12 of 20) and B (10 of 20), which hadreceived M. hyopneumoniae either concurrently with or prior toinoculation with PRRSV, respectively (Table 5). Isolation ofPRRSV from BAL peaked at day 10. Interestingly, at day 28,virus was isolated from only three of eight pigs in group A(concurrently infected with PRRSV and M. hyopneumoniae)and one of eight pigs in group B (infected with M. hyopneu-moniae prior to PRRSV), while all other pigs at that time werenegative for PRRSV.

FIG. 2. (a) Microscopic section of a normal lung from a pig. (b) Microscopic section of a lung from a pig infected 10 days previously with PRRSV. There is moderatediffuse interstitial pneumonia characterized by accumulation of necrotic debris and inflammatory cells in alveolar spaces, septal infiltration by mononuclear cells, andtype 2 pneumocyte hypertrophy and hyperplasia. (c) Microscopic section of a lung from a pig infected 28 days previously with M. hyopneumoniae. There isperibronchiolar and perivascular lymphoid hyperplasia characteristic of M. hyopneumoniae infection. (d) Microscopic section of a lung from a pig infected 28 dayspreviously with PRRSV and M. hyopneumoniae. Lesions characteristic of both M. hyopneumoniae- and PRRSV-induced pneumonia are present.

624 THACKER ET AL. J. CLIN. MICROBIOL.

on May 2, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

Titration of lung tissue for PRRSV revealed that the highestaverage titers of virus were found on day 10 in groups A, B, andF, with individual pig titers ranging from a TCID50 of 102 to105.6. The highest percentages of pigs with virus isolated fromthe lung homogenate were found in groups A and B; in each ofthese two groups, 83% of infected pigs had detectable PRRSVin lung tissue at necropsy. No differences were observed in thelevels of PRRSV between the various groups.

PRRSV antigen was detected in macrophages throughoutthe lungs of the PRRSV-infected pigs, as previously reported(12). The number of pigs which had PRRSV-positive cells wasgreatest at days 3 and 10 in groups A, B, and F (Table 5). Byday 28, only one of eight pigs in each of these groups hadPRRSV-positive macrophages. In group C, which had beeninoculated with PRRSV 10 days prior to the other groups, sixof six pigs were positive for PRRSV antigen by IHC at day 3,no positive cells were detectable at day 10, and three of eightpigs were positive by day 28. The six pigs in group C that werenecropsied at day 10 were all negative; however, PRRSV-likelesions were moderate and multifocal (microscopic score, 3.3of 6), and PRRSV was isolated from BAL of five of six pigs. Itis not uncommon for animals with subacute PRRSV to havelesions consistent with PRRSV yet to have negative IHC. IHCis not as sensitive as virus isolation from BAL fluids. No sta-

tistical differences in the number of IHC-positive cells wereobserved between the dually infected groups, A, B, and C, andgroup F, infected with PRRSV only. However, there was alarge variation in the number of positive cells among the indi-vidual pigs. In addition, no evidence of increased numbers ofcells containing PRRSV antigen was observed in the areas ofthe lungs with microscopic lesions consistent with M. hyopneu-moniae.

M. hyopneumoniae was isolated from all M. hyopneumoniae-inoculated groups. However, only 25% of pigs in group B(receiving M. hyopneumoniae on day 221 and PRRSV on day0) and 10% of pigs in group D (receiving M. hyopneumoniae onday 221) were positive for M. hyopneumoniae by culture (Ta-ble 6). Groups A (receiving PRRSV and M. hyopneumoniaeconcurrently) and C (receiving PRRSV on day 210) had thelargest numbers of pigs positive for M. hyopneumoniae. How-ever, no significant differences were observed in the mean titerof M. hyopneumoniae in lung tissue between the dually infectedgroups and the groups infected with M. hyopneumoniae alone(Table 6). All pigs in groups A, C, and E were positive for M.hyopneumoniae as determined by FA at day 28, while onlythree of eight and one of eight were positive in groups B andD, respectively.

TABLE 5. Isolation of PRRSV from BAL and titration of virus from lung tissue and number of cells with PRRSV antigen,as determined by IHC

Group

PRRSV result at the following necropsy date:

Day 3 Day 10 Day 28

BAL(no. pos.)a

IHC BAL(no. pos.)

IHC BAL(no. pos.)

IHC

No. pos.b No. of cellsc No. pos. No. of cells No. pos. No. of cells

A 3/6 4/6 101.8 6/6 3/6 91.5 3/8 1/8 87.3B 4/6 4/6 167 5/6 5/6 17.5 1/8 1/8 0.4Cd 2/6 6/6 8.2 5/6 0/6 0 0/8 3/8 208.9D 0/6 0/6 0 0/6 0/6 0 0/8 0/8 0E 0/6 0/6 0 0/6 0/6 0 0/8 0/8 0F 3/6 5/6 554 6/6 4/6 28.3 0/8 1/8 0.4G 0/6 0/6 0 0/6 0/6 0 0/7 0/8 0

a Number of pigs positive for PRRSV in BAL/total number of pigs in group. pos., positive.b Number of pigs with PRRSV antigen as detected by IHC/total number of pigs in group.c Average number of positive cells (mean number of positive cells in three lung sections) per pig with PRRSV antigen, as detected by IHC.d Group C received PRRSV on day 210 and M. hyopneumoniae on day 0 and so is not matched with the other groups with respect to the number of days after

inoculation with PRRSV.

TABLE 6. Prevalence of M. hyopneumoniae as determined by lung culture, titration of lung tissue, and FA

Group

M. hyopneumoniae result at the following necropsy date:

Day 3 Day 10 Day 28

No. of pigspositivea Titerb No. of pigs

FA positiveNo. of pigs

positive Titer No. of pigsFA positive

No. of pigspositive Titer No. of pigs

FA positive

A 5/6 105.0 0/6 5/6 106.0 4/6 8/8 106.6 8/8B 1/6 NDc 0/6 0/6 ND 0/6 3/8 ND 3/8Cd 3/6 106.0 1/6 5/6 107.2 6/6 8/8 107.0 8/8D 1/6 ND 0/6 1/6 ND 1/6 3/8 ND 1/8E 3/6 103.3 0/6 4/6 107.0 4/6 8/8 107.3 8/8F 0/6 ND 0/6 0/6 ND 0/6 0/8 ND 0/8G 0/6 ND 0/6 0/6 ND 0/6 0/7 ND 0/7

a Number of pigs positive for M. hyopneumoniae in ground lung tissue.b CCU of M. hyopneumoniae per milliliter.c ND, not done.d Group C received PRRSV on day 210 and M. hyopneumoniae on day 0 and so is not matched with the other groups with respect to number of days after inoculation

with PRRSV.

VOL. 37, 1999 M. HYOPNEUMONIAE AND PRRSV-INDUCED PNEUMONIA 625

on May 2, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

Serology. All pigs challenged with PRRSV developed anti-bodies to PRRSV by day 10 as determined by ELISA (data notshown). None of the pigs developed antibodies to M. hyopneu-moniae during the study, as expected from previous studies(23).

DISCUSSION

We have developed an experimental model demonstratingthat a mycoplasma can act as a cofactor in potentiating a viralpneumonia. Differentiating between the disease induced by thetwo pathogens was simplified by the pronounced differencesbetween their clinical manifestations and the resulting macro-scopic and microscopic lesions. PRRSV infection is character-ized by a severe, acute, diffuse, interstitial pneumonia, whereasM. hyopneumoniae induces a chronic, mild, localized broncho-pneumonia. These differences in pathologic lesions enabled usto identify the increased severity and duration of the PRRSV-induced pneumonia observed in all pigs infected with bothPRRSV and M. hyopneumoniae. M. hyopneumoniae potenti-ated the viral pneumonia independent of the time of viralinfection. Surprisingly, even when M. hyopneumoniae did notuniformly induce observable lesions typical of mycoplasmalpneumonia, as observed in group B, the viral pneumonia waspotentiated. This finding indicates that it is not just an additiveeffect of the different pathogens, but rather a potentiation ofPRRSV-induced pneumonia by M. hyopneumoniae. Specifi-cally, pigs in group B, which had received M. hyopneumoniae 21days prior to PRRSV inoculation, had a low level of mycoplas-mal pneumonia yet had PRRSV-induced lesions that persistedfor 4 weeks after PRRSV inoculation. PRRSV-induced lesionstypically are resolved by 4 weeks postinoculation, as confirmedby the group infected with PRRSV only (group F) (12).

PRRSV infection did appear to increase the severity of themacroscopic and microscopic M. hyopneumoniae-inducedpneumonia at day 10 in groups A and C, which receivedPRRSV concurrently with or prior to inoculation with M. hyo-pneumoniae. However, in both groups of pigs, the percentagesof lung tissue exhibiting macroscopic pneumonia consistentwith M. hyopneumoniae were not increased at day 28, which isour usual end point for studies of mycoplasmal pneumonia.Microscopically, however, PRRSV and mycoplasmal lesionswere more severe in groups A and C at day 28. These resultswere not observed in the pigs which were inoculated with M.hyopneumoniae prior to challenge with PRRSV (group B).These findings suggest that the presence of PRRSV early ininfection with M. hyopneumoniae may increase the rate atwhich pigs develop mycoplasmal pneumonia and increase thedamage induced at the cellular level but does not increase theoverall percentage of lung tissue exhibiting macroscopic pneu-monia.

Levels of either PRRSV or M. hyopneumoniae in tissue werenot increased in dually infected pigs. This suggests that theincreased severity and duration of PRRSV-induced pneumo-nia was not due to increased PRRSV or M. hyopneumoniaereplication. These results support the hypothesis that the in-flammatory response elicited in the course of mycoplasmalpneumonia may be a critical factor in the potentiation ofPRRSV-induced disease and lesions. M. hyopneumoniae-in-duced pneumonia is characterized by the infiltration of thelung parenchyma by mononuclear cells consisting primarily oflymphocytes and macrophages (26). In addition, it has beendemonstrated that M. hyopneumoniae has a mitogenic effect onswine lymphocytes (19). These findings suggest that activationof the immune system may contribute to the enhanced pneu-monia observed in our model. M. hyopneumoniae’s attraction

of these inflammatory cells may produce an ideal environmentfor PRRSV-induced inflammation to persist. These findingsare consistent with PRRSV infection of cells of the monocyte/macrophage/dendritic cell lineage, which is one of the majorcell types responding to M. hyopneumoniae infection. Thesteady influx of new monocytes/macrophages recruited by therelatively chronic M. hyopneumoniae infection may allowPRRSV to persist in the lungs at low levels for prolongedperiods.

The specific cellular and subcellular mechanisms by whichM. hyopneumoniae potentiates PRRSV are currently unknown.Recent studies conducted with humans have also identifiedmycoplasmas as potentially important cofactors in a number ofchronic disorders, including AIDS, malignant transformation,chronic fatigue syndrome, and various arthritides (15, 21, 27,29). These studies suggest that our findings are not unique orwithout precedent. The mechanisms by which mycoplasmasinfluence the pathogenesis of these diseases in humans havenot been elucidated. However, mycoplasmas are likely candi-dates for initiating disease because they produce chronic, oftenmild infections and have potent immunomodulatory proper-ties. Mycoplasmas could potentiate PRRSV by direct interac-tion with the virus, by interaction with cells from the immunesystem, or by inducing an inflammatory immune response bythe secretion of cytokines.

ACKNOWLEDGMENTS

We thank F. C. Minion for editorial assistance. We also thank N.Upchurch, B. Erickson, T. Young, R. Royer, T. Boettcher, and T.Anderson for technical assistance.

This work was supported by a grant from the National Pork Pro-ducers Council.

REFERENCES

1. Amanfu, W., C. N. Weng, R. F. Ross, and H. J. Barnes. 1984. Diagnosis ofmycoplasmal pneumoniae of swine: sequential study by direct immunofluo-rescence. Am. J. Vet. Res. 45:1349–1352.

2. Baseman, J. B., and J. G. Tully. 1997. Mycoplasmas: sophisticated, reemerg-ing, and burdened by their notoriety. Emerg. Infect. Dis. 3:21–32.

3. Bereiter, M., T. F. Young, H. S. Joo, and R. F. Ross. 1990. Evaluation of theELISA and comparison to the complement fixation test and radial immu-nodiffusion enzyme assay for detection of antibodies against Mycoplasmahyopneumoniae in swine serum. Vet. Microbiol. 25:177–192.

4. Blanchard, A., and L. Montagnier. 1994. AIDS-associated mycoplasmas.Annu. Rev. Microbiol. 48:687–712.

5. Collins, J. E., D. A. Benfield, W. T. Christianson, L. Harris, J. C. Hennings,D. P. Shaw, S. M. Goyal, S. McCullough, R. B. Morrison, H. S. Joo, D.Gorcyca, and D. Chladek. 1992. Isolation of swine infertility and respiratorysyndrome virus (isolate ATCC VR-2332) in North America and experimen-tal reproduction of the disease in gnotobiotic pigs. J. Vet. Diagn. Investig.4:117–126.

6. DeBey, M. C., C. D. Jacobson, and R. F. Ross. 1992. Histochemical andmorphologic changes of porcine airway epithelial cells in response to infec-tion with Mycoplasma hyopneumoniae. Am. J. Vet. Res. 53:1705–1710.

7. Del Giudice, R. A., N. F. Robillard, and T. R. Carski. 1967. Immunofluo-rescence identification of mycoplasma on agar by use of incident illumina-tion. J. Bacteriol. 93:1205–1209.

8. Duan, X., H. J. Nauwynck, and M. B. Pensaert. 1997. Virus quantificationand identification of cellular targets in the lungs and lymphoid tissues of pigsat different time intervals after inoculation with porcine reproductive andrespiratory syndrome virus (PRRSV). Vet. Microbiol. 56:9–19.

9. Galina, L. 1995. Possible mechanisms of viral-bacterial interaction in swine.Swine Health Production 3:9–14.

10. Halbur, P. G., J. J. Andrews, E. L. Huffman, P. S. Paul, X. J. Meng, and Y.Niyo. 1994. Development of a streptavidin-biotin immunoperoxidase proce-dure for the detection of porcine reproductive and respiratory syndromevirus antigen in porcine lung. J. Vet. Diagn. Investig. 6:254–257.

11. Halbur, P. G., M. F. Rothschild, B. J. Thacker, X.-J. Meng, P. S. Paul, andJ. D. Bruna. 1998. Differences in susceptibility of Duroc, Hampshire, andMeishan pigs to infection with a high virulence strain (VR2385) of porcinereproductive and respiratory syndrome virus (PRRSV). J. Anim. Breed.Genet. 115:181–189.

12. Halbur, P. G., P. S. Paul, M. L. Frey, J. Landgraf, K. Eernisse, X.-J. Meng,J. J. Andrews, M. A. Lum, and J. A. Rathje. 1996. Comparison of the antigen

626 THACKER ET AL. J. CLIN. MICROBIOL.

on May 2, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

distribution of two U.S. porcine reproductive and respiratory syndrome virusisolates with that of the Lelystad virus. Vet. Pathol. 33:159–170.

13. Halbur, P. G., P. S. Paul, M. L. Frey, J. Landgraf, K. Eernisse, X.-J. Meng,M. A. Lum, J. J. Andrews, and J. A. Rathje. 1995. Comparison of thepathogenicity of two U.S. porcine reproductive and respiratory syndromevirus isolates with that of the Lelystad virus. Vet. Pathol. 32:648–660.

14. Halbur, P. G., P. S. Paul, X.-J. Meng, M. A. Lum, J. J. Andrews, and J. A.Rathje. 1996. Comparative pathogenicity of nine U.S. porcine reproductiveand respiratory syndrome virus (PRRSV) isolates in a five-week-old cesar-ean-derived colostrum-deprived pig model. J. Vet. Diagn. Investig. 8:11–20.

15. Lo, S.-C. 1992. Mycoplasmas and AIDS, p. 525–545. J. Maniloff, R. N.McElhaney, L. R. Finch, and J. B. Baseman (ed.), In Mycoplasmas: molec-ular biology and pathogenesis. American Society for Microbiology, Wash-ington, D.C.

16. Lo, S. C., M. M. Hayes, R. Y. H. Wang, P. F. Pierce, H. Kotani, and J. W. K.Shih. 1991. Newly discovered mycoplasma isolated from patients infectedwith HIV. Lancet 338:1415–1418.

17. Meng, X.-J., P. S. Paul, P. G. Halbur, and M. A. Lum. 1996. Characterizationof a high-virulence U.S. isolate of porcine reproductive and respiratorysyndrome virus in a continuous cell line, ATCC CRL 11171. J. Vet. Diagn.Investig. 8:374–381.

18. Mengeling, W. L., K. M. Lager, and A. C. Vorwald. 1995. Diagnosis ofporcine reproductive and respiratory syndrome. J. Vet. Diagn. Investig.7:3–16.

19. Messier, S., and R. F. Ross. 1991. Interactions of Mycoplasma hyopneu-moniae membranes with porcine lymphocytes. Am. J. Vet. Res. 52:1497–1502.

20. Naot, Y. 1982. In vitro studies on the mitogenic activity of mycoplasmalspecies toward lymphocytes. Rev. Infect. Dis. 4:S205–S209.

21. Nicolson, G., and N. L. Nicolson. 1996. Diagnosis and treatment of myco-plasmal infections in Gulf War illness-CFIDS patients. Int. J. Occup. Med.Immunol. Toxicol. 5:69–78.

22. Ross, R. F. 1986. Mycoplasmal diseases, p. 469–483. In A. D. Leman, B.Straw, and R. D. Glock (ed.), Diseases of swine, 6th ed. Iowa State Univer-sity Press, Ames, Iowa.

23. Ross, R. F., B. J. Zimmermann-Erickson, and T. F. Young. 1984. Charac-teristics of protective activity of Mycoplasma hyopneumoniae vaccine. Am. J.Vet. Res. 45:1899–1905.

24. Simecka, J. W., S. E. Ross, G. H. Cassell, and J. K. Davis. 1993. Interactionsof mycoplasmas with B cells: antibody production and nonspecific effects.Clin. Infect. Dis. 17:S176–S182.

25. Stadtlander, C. T. K.-H., H. L. Watson, J. W. Simecka, and G. H. Cassell.1993. Cytopathogenicity of Mycoplasma fermentans (including strain incog-nitus). Clin. Infect. Dis. 17:289–301.

26. Suter, M., M. Kobisch, and J. Nicolet. 1985. Stimulation of immunoglobulin-containing cells and isotype-specific antibody response in Mycoplasma hyo-pneumoniae infection in specific-pathogen-free pigs. Infect. Immun. 49:615–620.

27. Taylor-Robinson, D. 1996. Mycoplasmas in rheumatoid arthritis and otherhuman arthritides. J. Clin. Pathol. 49:781–782.

28. Thanawongnuwech, R., E. L. Thacker, and P. G. Halbur. 1997. Effect ofporcine reproductive and respiratory syndrome virus (PRRSV) (isolateATCC VR-2385) infection on bactericidal activity of porcine intravascularmacrophages (PIMs): in vitro comparisons with pulmonary alveolar macro-phages (PAMs). Vet. Immunol. Immunopathol. 59:323–335.

29. Tsai, S., D. J. Wear, J. W.-K. Shih, and S.-C. Lo. 1995. Mycoplasmas andoncogenesis: persistent infection and multistage malignant transformation.Proc. Natl. Acad. Sci. USA 92:10197–10201.

VOL. 37, 1999 M. HYOPNEUMONIAE AND PRRSV-INDUCED PNEUMONIA 627

on May 2, 2020 by guest

http://jcm.asm

.org/D

ownloaded from