Effect of Genetic and Proteomic Variation in Field Isolates of Mycoplasma Hyopneumoniae on Virulence, Diagnosis and Prevention O

Iowa State University Capstones, Theses and Retrospective Theses and Dissertations Dissertations

2008 Effect of genetic and proteomic variation in field isolates of Mycoplasma hyopneumoniae on virulence, diagnosis and prevention of disease in the pig Erin Linn Strait Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/rtd Part of the Agriculture Commons, Animal Sciences Commons, Large or Food Animal and Equine Medicine Commons, Microbiology Commons, and the Veterinary Pathology and Pathobiology Commons

Recommended Citation Strait, Erin Linn, "Effect of genetic and proteomic variation in field isolates of Mycoplasma hyopneumoniae on virulence, diagnosis and prevention of disease in the pig" (2008). Retrospective Theses and Dissertations. 15640. https://lib.dr.iastate.edu/rtd/15640

This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected].

Effect of genetic and proteomic variation in field isolates of Mycoplasma hyopneumoniae on virulence, diagnosis and prevention of disease in the pig

Erin Linn Strait

A dissertation submitted to the graduate faculty

in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

Major: Immunobiology

Program of Study Committee: Eileen L. Thacker, Major Professor Patrick J. Halbur Douglas E. Jones F. Chris Minion Brad J. Thacker

Iowa State University

Ames, Iowa

2008

3307067 2008 ii

TABLE OF CONTENTS

LIST OF FIGURES vi LIST OF TABLES vii ABSTRACT ix CHAPTER 1. GENERAL INTRODUCTION 1 Introduction 1 Dissertation Organization 1 Literature Review 2 Mollicutes 2 Mycoplasmas and Acholeplasmas of swine 3 Mycoplasma hyopneumoniae 3 Disease characteristics 3 Epidemiology 4 Field isolate diversity 5 Virulence factors 6 Immune response 7 Vaccines 8 Diagnostics 9 Culture 9 Serology 9 Polymerase chain reaction 10 Summary with Hypothesis and Goals 11 References 12

CHAPTER 2. COMPARISON OF MYCOPLASMA HYOPNEUMONIAE FIELD ISOLATES 22 Abstract 22 Introduction 22 Materials and Methods 24 Mycoplasma hyopneumoniae field isolates 24 Study design 24 Sample collection 25 Serology 25 iii

BALF analysis 25 Growth curves 26 Statistics 26 Results 26 Challenge inoculum titers 26 Percent pneumonia 27 Bacterial isolation 27 Quantitative real-time PCR 29 IL-1ß and IL-18 ELISAs 29 IgA and IgG ELISAs 29 Serology 29 Growth curves 29 Discussion 31 Acknowledgements 32 References 33

CHAPTER 3. IMPACT OF GENETIC DIVERSITY AMONG MYCOPLASMA HYOPNEUMONIAE FIELD ISOLATES ON PCR ASSAY DEVELOPMENT 35 Abstract 35 Introduction 35 Materials and Methods 36 Mycoplasmas 36 PCR assays 36 M. hyopneumoniae mhp165 real-time PCR 37 M. hyopneumoniae mhp183 real-time PCR 43 mhp165 and mhp183 multiplex real-time PCR assay 43 DNA sequencing and analysis 43 Results 45 Mycoplasma PCR assay panel 45 Real-time PCR assays 45 DNA sequence results 45 Detection level 46 Discussion 47 Acknowledgements 49 iv

References 50

CHAPTER 4. COMPARISON OF MYCOPLASMA HYOPNEUMONIAE- SPECIFIC ELISAS AND THE EFFECT OF FIELD ISOLATE VARIABILITY 53 Abstract 53 Introduction 54 Materials and Methods 55 Animals, housing and experimental design 55 Sample collection 55 Macroscopic pneumonia 56 ELISA assays 56 PCR from nasal swabs, bronchial swabs and bronchial alveolar lavage fluid 57 Western blot assay 57 Statistics 57 Results 57 Macroscopic pneumonia 57 Bacteriological findings 58 Oxoid, IDEXX and 232 Tween 20 ELISAs 58 97MP0001 Tween 20 ELISA 62 PCR 63 Western blot 63 Discussion 64 Acknowledgements 68 References 68

CHAPTER 5. EFFICACY OF A MYCOPLASMA HYOPNEUMONIAE BACTERIN IN PIGS CHALLENGED WITH TWO CONTEMPORARY PATHOGENIC ISOLATES OF M. HYOPNEUMONIAE PROVIDES AN INNOVATIVE APPROACH TO EVALUATING M. HYOPNEUMONIAE VACCINES 71 Summary 71 Introduction 72 Materials and Methods 73 Study animals 73 v

Experimental design 73 Serum antibody tests 74 Necropsy 74 Data analysis 75 Results 76 Death of animals during the study 76 Serology 76 Macroscopic lung lesions 76 M. hyopneumoniae real-time PCR 76 IgA and IgG antibodies 78 Bacterial isolation from bronchial swabs 79 Discussion 81 Implications 83 Acknowledgements 83 References 83

CHAPTER 6. GENERAL CONCLUSIONS 87 General Discussion 87 Recommendations for Future Research 89

ACKNOWLEDGEMENTS 90

LIST OF FIGURES

Figure 2.1. Box-plot comparing the percent of macroscopic pneumonia in pigs challenged with one of four field isolates of M. hyopneumoniae, strain 232, or unchallenged controls 27

Figure 2.2. Growth-curves for four M. hyopneumoniae field isolates and strain 232 generated from real-time PCR on culture samples 30

Figure 3.1. Sequence alignment of the mhp165 gene real-time PCR target region for isolates of M. hyopneumoniae compared to the consensus sequence from M. hyopneumoniae strain 232 (AE017332) 42

Figure 3.2. Sequence alignment of the mhp183 gene real-time PCR target region for isolates of M. hyopneumoniae compared to the consensus sequence from M. hyopneumoniae strain 232 (AE017332) 44

Figure 3.3. DNA sequence analysis of the Kurth et al. inner PCR product of strains 00MP1502 and 00MP1301 47

Figure 4.1. 232 Tween 20 OD comparison of group means in M. hyopneumoniae-infected groups versus days-post-challenge 61

Figure 4.2. 97MP0001 Tween 20 OD comparison of group means in M. hyopneumoniae-infected groups versus days-post-challenge 63

Figure 4.3. Western blot results detecting proteins from antigen prepared from M. hyopneumoniae isolate 97MP0001 64

vii

LIST OF TABLES

Table 2.1. Summary of pneumonia and bronchial alveolar lavage fluid analysis 28

Table 2.2. A comparison of Tween 20, IDEXX and Oxoid ELISA results from sera collected at necropsy (28 DPC) from pigs challenged with one of four field isolates of M. hyopneumoniae or strain 232 30

Table 3.1. Bacteria used in this study 38

Table 3.2. Summary of results for PCR assays tested against M. hyopneumoniae isolates 41

Table 3.3. List of primer and probe sequences used in this study 42

Table 3.4. Analysis of clinical samples by real-time PCR 46

Table 4.1. Summary of the percentage of pneumonia, culture, and PCR results or pigs challenged with M. hyopneumoniae field isolates 59

Table 4.2. Summary of pneumonia and culture results for pigs challenged with M. flocculare, M. hyorhinis and M. hyosynoviae 60

Table 4.3. ELISA results from pigs challenged with M. hyopneumoniae field isolates 62

Table 5.1. Frequency of pigs positive for M hyo Tween 20 and DAKO serum antibody titers in pigs vaccinated twice with an M hyo bacterin, or saline, then challenged with M hyo, and necropsied 28 DPC (95MP1509 challenge) or 30DPC (00MP1301 challenge) 77

Table 5.3. Parametric analysis of M hyo percent lung lesions in pigs vaccinated twice with an M hyo bacterin, or saline, then challenged with M hyo, and necropsied 28 DPC (95MP1509 challenge) or 30DPC (00MP1301 challenge) 79

viii

Table 5.5. M hyo real-time polymerase chain reaction levels in bronchial alveolar lavage fluids (BALF) at necropsy in pigs vaccinated twice with an M hyo bacterin, or saline, then challenged with M hyo, and necropsied 28 DPC (95MP1509 challenge) or 30DPC (00MP1301 challenge) 80

Table 5.6. M hyo-specific IgA and IgG antibodies in bronchial alveolar lavage fluids (BALF) at necropsy in pigs vaccinated twice with an M hyo bacterin, or saline, then challenged with M hyo, and necropsied 28 DPC (95MP1509 challenge) or 30DPC (00MP1301 challenge) 81

ABSTRACT

The purpose of these studies was to investigate the implications of variability in virulence, genetics and antigenic profiles among field isolates of Mycoplasma hyopneumoniae. Increased virulence was compared with in vitro and in vivo growth characteristics of the organism and with immune parameters in the pig, although none was found to be significantly correlated. The impact of proteomic and genetic variation among field isolates on diagnostic assays was examined by both in vivo and in vitro studies. Three enzyme-linked immunosorbent assays (ELISAs) used in the United States were compared and shown to not detect all isolates of M. hyopneumoniae equally. Similarly, polymerase chain reaction (PCR) assays were tested against a panel of M. hyopneumoniae field isolates and it was shown that some isolates are not detected by all published PCR assays. Therefore, two new real-time PCR assays were developed that appear to be both sensitive and specific for M. hyopneumoniae. The ability of a commercial vaccine to protect against field isolates of M. hyopneumoniae was evaluated in a challenge study using two recent field isolates. One isolate produced nearly twice as many macroscopic lung lesions as the other isolate, but vaccinated pigs had less pneumonia compared to non-vaccinated pigs in both challenge groups. Together these results demonstrate that differences exist among isolates of M. hyopneumoniae in the field at the genetic and antigenic level and that these differences impact the virulence of the organism, diagnostic capabilities in the pig, and the prevention of M. hyopneumoniae-associated disease.

CHAPTER 1. GENERAL INTRODUCTION

Introduction Mycoplasma hyopneumoniae, the causative agent of enzootic pneumonia of pigs, is among the most significant pathogens of swine today leading to economic losses for producers world-wide. Mycoplasma hyopneumoniae is also a key component of the porcine respiratory disease complex which further increases the impact of this organism in pigs through enhanced pneumonia by potentiating and prolonging the disease caused by other organisms such as porcine reproductive and respiratory syndrome virus (PRRSV) and porcine circovirus type 2 (PCV2). Although M. hyopneumoniae was first described in 1965, relatively little is known about this organism due to its fastidious growth characteristics and unique genetic features that impede the use of routine genetic manipulations commonly employed to assign gene functions. In addition, while differences in protein expression and gene expression among isolates of M. hyopneumoniae have been identified, little is understood concerning the implications for virulence, therapeutics and diagnostics that this variation may have. The studies performed here investigated the impact that different field isolates of M. hyopneumoniae had on these factors.

Dissertation Organization This dissertation is organized into six chapters. Chapter one is the introduction to M. hyopneumoniae. The next four chapters describe research performed that have been or will be submitted for publication. Chapter two investigated the differences in virulence between field isolates of M. hyopneumoniae. Chapters three and four describe the impact that genetic and protein-expression differences among field isolates of M. hyopneumoniae have on the ability of commonly used diagnostic procedures to detect the organism. Chapter five measures the ability of a commercial vaccine to protect against two virulent field isolates of M. hyopneumoniae. A general conclusion and future studies comprise Chapter six. The presentation of chapters two through five follow the style of the journal to which they are being or were submitted. 2

Literature Review

Mollicutes The class Mollicutes includes eight genera and over 200 members (http://www.the- icsp.org/subcoms/mollicutes.htm). Organisms in this class are pleomorphic due to the lack of a cell wall (Mollicutes means ‘soft skin’) and the presence of an internal cytoskeleton in some species (70, 71). Mollicutes are the smallest self-replicating organisms known and have genomes ranging from 577-2220 kilobase pairs (69). They are filterable organisms with diameters ranging from 0.3-1 µm (69). It is thought that they attained their small genomes through degenerative evolution from a gram-positive ancestor (43). During this process they lost a number of important metabolic pathways including those for phospholipid, amino acid, and purine and pyrimidine synthesis. The Mollicutes are also unique in that their genomes contain a high A+T content from 60-77 mol% (32) and they utilize UGA codons to encode tryptophan rather than a stop codon (58, 104). While capable of translating all available codons, they possess a greatly reduced number of tRNAs (around 20 of the 62 potential tRNAs) (58). This combination of traits makes members of the class Mollicutes highly dependent on their host for growth and survival, and also highly susceptible to environmental conditions. Mollicutes are pathogens and commensals of a diverse variety of hosts including both animals and plants (and their insect vectors). Many of the mammalian Mollicutes reside on mucosal surfaces of the respiratory or urogenital tract of their hosts, or in joint spaces (83). Most of these organisms are extra-cellular, but evidence exists for some Mollicutes residing intracellularly (7, 77). A few members are blood-borne (50). Most clinical isolations of species in the class Mollicutes belong to the genus Mycoplasma (34). Many of the Mycoplasmas are slow-growing and nutritionally demanding in culture, requiring enriched media. Therefore, the true significance of most Mycoplasmas is currently under appreciated because of low rates of isolation and the lack of reliable detection methods for these organisms (34). In addition, surveys of tissue cell cultures have found that from 15-80% are contaminated with Mollicutes (98). As many as 20 different Mollicute species have been found in tissue culture systems with Mycoplasma arginini, Mycoplasma fermentans, Mycoplasma hyorhinis, Mycoplasma orale and Acholeplasma laidlawii accounting for 95% of these occurrances (63, 95). 3

Mycoplasmas and Acholeplasmas of swine While M. hyopneumoniae is the most notable mycoplasma infecting swine, a number of other members of the class Mollicutes have also been identified in pigs. In the respiratory tract, Mycoplasma hyorhinis and Mycoplasma flocculare are commonly isolated with Mycoplasma hyosynoviae being isolated occasionally. None of these mycoplasmas (aside from M. hyopneumoniae) are considered pathogens when found in the respiratory tract, although M. hyorhinis and M. hyosynoviae are capable of inducing arthritis (67, 68). Mycoplasma suis is a blood-borne organism that until recently was classified as Eperythrozoon suis (59, 60). It adheres to the surface of erythrocyte membranes and induces anemia (84). Acholeplasma laidlawii and Acholeplasma granularum have both been isolated from clinical samples from pigs, but their contribution to disease is unknown at this time (26). Mycoplasma hyopneumoniae has a genome size of 892 kilobase pairs and one of the highest A+T contents among the Mycoplasmas (71%) (55). Mycoplasma hyopneumoniae is one of the most fastidious mycoplasmas to culture. It requires a highly-enriched media and is easily overgrown by other species of mycoplasmas and other types of bacteria that also inhabit the respiratory tract (24). In addition, it does not grow readily on agar making traditional titering methods impossible. Therefore, a less-precise method based on color- changing units (CCU) is usually employed (91). Mycoplasma hyopneumoniae’s growth characteristics also make performing minimum inhibitory concentration (MIC) assays testing for antibiotic sensitivity tedious and cumbersome (28). New methods using flow cytometry have been described to detect mycoplasmas, including M. hyopneumoniae, which may greatly enhance our ability to quantify these organisms (5). Three strains of M. hyopneumoniae: 232, J and 7448 have been completely sequenced (55, 99). While sequencing of M. hyopneumoniae has provided important information about its genome, over half (56%) of the gene sequences identified are unique to M. hyopneumoniae (55). Currently gene deletion experiments commonly used to assign functions to genes in other organisms cannot be performed because of M. hyopneumoniae’s unique genetics and poor growth characteristics on agar. As a result, our knowledge of the function of most of M. hyopneumoniae’s genes is lacking.

Mycoplasma hyopneumoniae

Disease characteristics Mycoplasma hyopneumoniae is the causative agent of enzootic pneumonia and is found in most pig-producing countries worldwide. It induces a chronic pneumonia with 4 high morbidity and low mortality with the most notable clinical sign being a dry, non- productive cough (73, 85, 101). Mycoplasma hyopneumoniae colonizes the ciliated epithelium of the pig respiratory tract (10, 78, 107), but in specific pathogen-free pigs has been demonstrated to spread systemically (possibly through the blood and/or lymphatics) in the early phase of infection (44). After M. hyopneumoniae attaches to the ciliated respiratory epithelium it induces clumping of the cilia through an unknown mechanism (18) leading to impairment of the mucociliary apparatus. This is thought to predispose the pig to secondary infections. Macroscopic lesions appear primarily cranial-ventrally as well-demarcated, tan-to-purple areas of lung consolidation. Microscopically M. hyopneumoniae infection is characterized by peribronchiolar, peribronchial, and perivascular areas of lymphoid hyperplasia with mononuclear infiltrates into the affected airways. Both the macroscopic and microscopic lesions are characteristic, but not pathognomonic of M. hyopneumoniae infection. Factors affecting the severity of disease include, but are not limited to environmental conditions (cleanliness of the environment including air quality), presence of other pathogens, stocking density, production system (continuous flow versus all-in/all-out), and vaccination status (42, 81, 101). Mycoplasma hyopneumoniae-induced pneumonia takes months to resolve and pigs can carry the organism long-term (12, 85). Although M. hyopneumoniae is a primary pathogen, it is more significant as a member of the porcine respiratory disease complex (PRDC). This complex consists of viruses including PRRSV, PCV2, swine influenza virus (SIV) and bacteria such as Pasturella multocida, Bordetella bronchiseptica, Streptococcus suis, along with M. hyopneumoniae. Of the pathogens associated with the PRDC, M. hyopneumoniae is recognized as one of the most commonly isolated and most clinically important. Mycoplasma hyopneumoniae contributes not only additively to the disease complex, but has been demonstrated to potentiate and prolong the disease caused by both PRRSV and PCV2 (62, 96). The PRDC is the most significant source of economic loss in the swine industry today, but little is known about the mechanisms of pathogenesis or interactions between the organisms involved.

Epidemiology The most important mode of transmission of M. hyopneumoniae to a naïve herd is through introduction of a subclinically infected carrier pig to the system (53, 65). There is a relatively high risk of this occurring because M. hyopneumoniae often spreads slowly within a herd leading to a gradual onset of disease in the population. Airborne transmission has been demonstrated, but is considered a less important source of 5 infection as M. hyopneumoniae is relatively labile in the environment. Transmission by this route is more likely to occur in humid, windy conditions (27). Dry, freezing, and rainy weather are negatively associated with spread of M. hyopneumoniae (88). Within a herd, spread occurs through both vertical and horizontal means. Piglets can be infected from their sows, who may actively shed the organism at the time of parturition and lactation (12, 23, 82). Maternal antibodies provide some protection from severity of disease, but do not prevent colonization from occurring in the piglets (45, 103). Pigs that are infected prior to weaning go on to spread it to other pigs when litters are mixed at the grow-finish stage (101).

Field isolate diversity Much research highlighting differences between field isolates of M. hyopneumoniae exists. Studies using SDS-PAGE alone, or in combination with western blot immunoassays have been used extensively. Considerable variation in banding patterns among a number of isolates has been demonstrated using total protein and glycosylated protein-only staining techniques, or reacting with convalescent sera or monoclonal antibodies (6, 39, 79). Each of these assays identified differences in expression of several proteins among the isolates tested. Antigenic diversity has also been demonstrated using antisera raised against field isolates of M. hyopneumoniae in growth inhibition, metabolic inhibition and 2-dimensional immunoelectrophoresis assays (72). Comparison of the three genomes of M. hyopneumoniae that have been sequenced provides evidence of intra-specific rearrangements in each of these strains (99). Unlike other mycoplasma species, M. hyopneumoniae has relatively few tandem repeats within gene coding sequences that impact size or expression of surface proteins (55). However, amplification of DNA sequences containing variable numbers of tandem repeats (VNTR) has been used to differentiate between field isolates (17). Another study evaluating molecular techniques including amplified fragment length polymorphism analysis, random amplified polymorphic DNA (RAPD) analysis, restriction fragment length polymorphism analysis and VNTR analysis were able to distinguish between isolates of M. hyopneumoniae from an international collection (87). Multilocus sequence typing has also been used for typing field isolates of M. hyopneumoniae (48). Although a considerable amount of research has been conducted to show differences between field isolates of M. hyopneumoniae, especially at the levels of protein and gene expression, few studies have been able to correlate any of these changes with differences in virulence. Vicca, et al. found a link between a 5000 bp fragment and increased virulence in a small number of isolates using RAPD analysis (102). Another study 6 investigating two isolates from the Vicca, et al. study demonstrated that a high-virulent isolate had quicker generation times in vitro than did a low-virulent isolate suggesting a correlation of virulence to this trait (52).

Virulence factors In the respiratory tract M. hyopneumoniae attaches to the ciliated respiratory epithelium. This association has been shown to be required for virulence, and that the protein P97 is involved in this process (29). P97 is part of an operon with two open- reading frames, P97 and P102 (31). The P97 structural gene has two repeat regions, R1 and R2, and monoclonal antibodies that block attachment of M. hyopneumoniae to the cilia in vitro (105) have been shown to bind to the R1 region (30, 54). Minion et al. also showed that eight or more repeats within the R1 region was sufficient for cilium binding (54). Additional potential adhesins have been identified (11, 15) (GenBank AF279908), but their role in binding has not been confirmed. Attachment of M. hyopneumoniae to the respiratory epithelium leads to clumping of the cilia and ciliostasis in these cells and ultimately to loss of the cilia. This phenomenon has been demonstrated in vitro using tracheal ring cultures, but the mechanism through which this occurs remains to be elucidated (18). Intracellular free-calcium release by the epithelial cells has been correlated with this process (64), and a 54 kDa membrane protein shown to be cytopathic to human lung fibroblasts in culture may be involved (25). Mycoplasma hyopneumoniae is surrounded by a capsular matrix. Although no clear connection has been made, a thinner capsule is produced after serial passage in vitro, and this was correlated with decreased pathogenicity in vivo (93). It is unclear what other effects occurred through serial passage that also may have affected virulence. Because M. hyopneumoniae lacks the ability to synthesize many of its own building blocks for growth, the organism requires mechanisms for the acquisition of molecules such as cholesterol, phospholipids, purines, pyrimidines, and amino acids. The means by which M. hyopneumoniae extracts these macromolecular precursors from the host are largely undefined but are generally thought to involve some means of host-tissue destruction that contributes to pathogenesis. A 54 kDa protein may be involved directly in causing cell death, and/or the effect may be indirect, caused by the immune response of the pig to infection (25). Neutrophils are a prominent cell type in infected airways, and these cells are known to release superoxide radicals and other toxic substances that kill not only pathogens, but also host cells. In addition, lipoproteins have been isolated from several species of mycoplasma including M. fermentans, M. pneumoniae, and M. arthritidis that are inflammatory in 7 nature and may act as non-specific mitogens (21, 36, 37). While no specific lipoprotein from M. hyopneumoniae has been identified with these capabilities, there is data to suggest that at least one exists. One study found that lymphocytes are non-specifically stimulated by unidentified membrane constituents of M. hyopneumoniae (51). Over 50 lipoproteins have been predicted through sequence analysis of the genome (55), and a microarray of M. hyopneumoniae genes demonstrated that transcription of three of these predicted lipoproteins are upregulated during heat-shock conditions (a measure of the stress response) in vitro (40). This non-specific stimulation may be the cause of the peribronchiolar, peribronchial and perivascular lymphoid hyperplasia, leading to an ineffective immune response and tissue damage that likely benefits the growth of M. hyopneumoniae.

Immune response Immunopathologic changes are a major component of M. hyopneumoniae-induced disease, although little is understood about the mechanisms underlying this process. Interestingly, thymectimized pigs treated with anti-thymocyte serum displayed decreased lesion severity in response to experimental M. hyopneumoniae infection (94) suggesting the cell-mediated immune responses may actually contribute to the pathology observed during infection. This study also found that the thymectomized/anti-thymocyte treated pigs infected with M. hyopneumoniae had a greater tendency for systemic spread of M. hyopneumoniae compared to pigs that only received M. hyopneumoniae challenge. In addition, the thymectomized/anti-thymocyte treated pigs infected with M. hyopneumoniae had increased intra-luminal exudate compared to pigs that only received the M. hyopneumoniae challenge, and the cell types comprising the exudate were different between the two groups. In the thymectomized/anti-thymocyte treated pigs the predominant intra-luminal cells were large mononuclear cells, whereas in the group that only received M. hyopneumoniae, most of the cells were polymorphonuclear leukocytes. The importance of the systemic spread and differences in the cellular response is unknown. Mucosal epithelial cells have a critical role in transmitting signals between microbial pathogens and the underlying cells in the mucosa (33). While M. hyopneumoniae binds only to the extracellular matrix associated with the cilia (107), there is evidence that these epithelial cells respond to M. hyopneumoniae binding. When M. hyopneumoniae attaches to cilia, increases in the intracellular concentration of calcium are observed in the epithelium (64). This increase in calcium suggests the involvement of G-protein mediated cell signaling leading to upregulation of transcription factors and ultimately cell 8 activation. This activation likely influences the local milieu of cytokines and chemokines secreted at the site of infection, which are critical for determining the outcome of both the innate and adaptive immune responses. Toll-like receptors (TLR) may be an important link between the previously discussed lipoproteins in the membrane of M. hyopneumoniae and the inflammation observed during infection. Specifically, TLR2 and TLR6 have been shown to mediate TNFα expression after infection with M. hyopneumoniae (57). Toll-like receptors 2 and 6 are expressed by macrophages and mediate an inflammatory response by initiating an intracellular signaling cascade that culminates in the transcription of inflammatory cytokines. While TLR4 binds to lipopolysaccarides (LPS), TLR2 and TLR6 activate the same signaling cascade by binding lipoproteins. Interleukin (IL)12 and IL10 have been associated with Th1 and Th2 lymphocyte responses, respectively. Increased levels of both cytokines have been measured in the lungs of M. hyopneumoniae-infected pigs (74, 97). To protect against M. hyopneumoniae, an extracellular pathogen, it is thought that a controlled Th1 response in the lungs would be the most effective immune response, but the cytokine profile after infection is mixed. Production of proinflammatory cytokines IL1, IL2, IL8, IL12, IL18 and TNFα suggest a Th1 response, but the presence of increased levels of IL4, IL6 and IL10 support a Th2 response (3, 4, 16, 56, 74, 75). This suggests the presence of a conflicting and ineffective immune response. The lack of immune-related reagents and an in-depth understanding of the pig immune system have hampered attempts to understand the pathogenesis of M. hyopneumoniae and other disease-causing agents in swine. In recent years a significant portion of the sus scrofa genome has been sequenced and annotated allowing for more detailed studies. Microarrays are a relatively new technology, which can be used to screen samples for up- or down-regulation of large numbers of genes simultaneously. An array of swine genes has been developed, but is based on muscle tissue so few immune- related genes were included (106).

Vaccines The commercial M. hyopneumoniae vaccines currently on the market are all killed whole-cell or membrane preparations with a variety of adjuvants. No modified-live vaccines exist. These bacterins have been shown to decrease mycoplasmal pneumonia but do not prevent colonization of the host, and not all preparation methods have been shown to be effective. Recently, increased numbers of vaccine failures have been reported. This may be due to genetic variability in M. hyopneumoniae field strains (41). One study comparing various vaccine preparations found that extracts prepared according to a 9 freeze-thaw procedure were variable in protective activity and, in some instances, enhanced lesion development (76). It is presently unknown what specific components of M. hyopneumoniae are important to include (or possibly exclude) in a vaccine in order to provide the maximum protection. Experimental studies using sub-unit or DNA vaccines have shown some protection against pneumonia, but none were more effective than whole-cell bacterins (22, 35, 80). One explanation for this is that single-antigen vaccines are less-capable of providing protection against complex bacterial pathogens. Currently available commercial vaccines increase the local and systemic levels of antibodies. However, there is pig-to-pig variation in the magnitude of the immune response observed in each pig that does not correlate well with protection. As with factors surrounding lesion formation, the protective mechanism by which these vaccines act to decrease pneumonia has yet to be defined. To date, most challenge studies testing the efficacy of M. hyopneumoniae vaccines have relied on the use of a few, well- characterized strains. The effect on vaccine efficacy of the recently reported differences in virulence among field isolates of M. hyopneumoniae has not been determined.

Diagostics

Culture Typically, isolation and culture of an organism is considered the “gold” standard for diagnosis and confirmation of its presence in the host. However, isolation of M. hyopneumoniae is impeded by a number of factors. This organism is nutritionally demanding and is cultured in a highly enriched growth media that requires M. hyopneumoniae antibody-negative serum (24). In addition, primary isolation commonly requires four to eight weeks to reach measurable levels in culture. Contamination from other bacteria and mycoplasmas that grow more readily in the media frequently preclude attempts to obtain pure cultures of the organism. Furthermore, once a mycoplasma culture is obtained, species identification is required using various techniques such as polymerase chase reaction (PCR), immunofluorescence, or immunohistochemistry, although antibodies for these assays are not commercially available (62, 66, 86).

Serology Serology is the most common tool used to indirectly determine the presence or absence of an organism within a herd due to its low cost and the ease of sample collection. However, as with most diagnostic assays associated with M. hyopneumoniae infection, interpretation of serological results may be challenging. Currently, many 10 producers are interested in eliminating M. hyopneumoniae from their herds, but several characteristics of M. hyopneumoniae make this difficult to achieve and maintain. Seroconversion after infection often takes four to eight weeks to achieve. When this trait is combined with the often slow spread of M. hyopneumoniae within a herd it makes accurate and timely diagnosis within individual animals difficult. This complicates the use of serology as a reliable screen for replacement animals. Current M. hyopneumoniae ELISAs include the Tween-20 ELISA (8, 61) and the HerdChek® M. hyopneumoniae Antibody Test Kit (IDEXX Laboratories, Westbrook, ME), which are both indirect ELISAs. The IDEIA™ M. hyopneumoniae EIA KIT (Oxoid, Ely, Cambridgeshire, UK, which is formerly the DAKO M. hyopneumoniae ELISA) is a competitive-inhibition ELISA. The antibodies detected in the indirect assays recognize numerous proteins primarily associated with the cell membrane, while the Oxoid ELISA is designed to detect antibodies to a single intracellular protein of M. hyopneumoniae. Each of these strategies has advantages and limitations. Antigenic variability between isolates of M. hyopneumoniae has been documented (6, 14) and recent studies indicate that the current ELISAs vary in their ability to detect antibodies from pigs experimentally infected with different field isolates of M. hyopneumoniae (92, 101). Using western blot analysis, Ameri, et al. found pigs in the field develop different antibody profiles after natural infection with M. hyopneumoniae (1). In his study, Ameri divided the sera into categories including samples negative by the competitive inhibition ELISA (DAKO), but positive by an indirect ELISA (IDEXX), and vice versa. The findings from this study suggest that not all pigs produce antibodies to the single protein used in the competitive inhibition ELISA. An assay based on more than one protein may decrease the likelihood of false-negative results. Previous studies have found both the competitive-inhibition and indirect ELISAs to be highly specific to M. hyopneumoniae (20, 85) although false-positive results from sera with a high titer to M. flocculare by the Tween 20 indirect ELISA has been described (8). These findings increase concerns of potential false-positive results in the indirect ELISAs which use whole-cell extracts of cultured M. hyopneumoniae as antigen.

Polymerase-chain reaction. A number of PCR assays have been developed and reported to be both sensitive and specific for M. hyopneumoniae (9, 13, 19, 38, 46, 85, 86, 89, 90, 100). As a result, this technique is now among the most widely used for detection of M. hyopneumoniae in pigs. While the development of PCR assays has greatly enhanced our ability to detect M. hyopneumoniae, genetic variability among different isolates has been demonstrated (2, 11

17, 41, 47, 49, 87). It is not known to what extent this heterogeneity affects the sensitivity of the different assays. Even though the genome of three different strains of M. hyopneumoniae have been sequenced (55, 99), there is little data available concerning which genes are conserved among M. hyopneumoniae field isolates. A recent study found that in order to increase the detection of M. hyopneumoniae on Swiss farms, two PCR assays specific for different genes were required in order to detect all of their M. hyopneumoniae field samples (19). Mycoplasma hyopneumoniae colonizes the lower respiratory tract; therefore, nasal swabs for PCR analysis are only moderately sensitive unless the pigs are actively shedding the organism through coughing. One study evaluated the use of polyethersulfone membranes (pore size, 0.2 micron) mounted in filter holders for air sampling in swine confinement buildings (89). Under field conditions, the sampling system was able to detect airborne M. hyopneumoniae on 80% of farms where acute respiratory disease was present. No airborne M. hyopneumoniae was detected on infected farms that did not have acute cases. Samples obtained from bronchoalveolar lavage fluid and tracheobronchial sites have been shown to be the most predictive of infection (38).

Summary with Hypothesis and Goals While M. hyopneumoniae remains as an important and wide-spread pathogen in today’s swine industry, little is understood in the way of virulence factors used by the organism or protective mechanisms needed in the host. It is also unknown what impact variation between field isolates at the genetic and protein expression levels have on virulence, detection or protection. This is largely due to the fastidious nature and unique genetics of M. hyopneumoniae that have hindered detailed study of the organism. The overall hypothesis of this study is that field isolates of M. hyopneumoniae that are distinct by genetic and/or protein analysis possess a range of virulence and will not be equally detected through PCR or ELISA, nor will current vaccines be equally protective. Therefore the goal of the research performed in this group of studies is to evaluate the ability of different isolates of M. hyopneumoniae to cause disease and also to determine if currently used diagnostic assays and protective measures are affected by differences between isolates of M. hyopneumoniae from the field. The first experiment described in Chapter 2 involves challenging pigs with four different field isolates of M. hyopneumoniae and comparing them to pigs infected with a traditional challenge strain, 232. Macroscopic pneumonia was measured in each pig at necropsy and compared between groups. In addition, growth characteristics of the 12 isolates in vitro and in vivo and cytokine expression were analyzed and correlated to virulence. The next two chapters evaluated whether genetic or antigenic variation within isolates of M. hyopneumoniae affected the detection ability of currently used diagnostic assays. Chapter 3 evaluated the ability of currently used M. hyopneumoniae-specific PCR assays to detect a panel of M. hyopneumoniae isolates. In addition, two new real-time PCR assays were developed that appear to be sensitive and specific for M. hyopneumoniae. In Chapter 4 a time-course study was undertaken to acquire serum from pigs challenged with four different isolates of M. hyopneumoniae, two isolates of M. flocculare and one isolate each of M. hyorhinis and M. hyosynoviae in order to test the sensitivity and specificity of currently used M. hyopneumoniae-specific ELISAs. In Chapter 5 pigs were either vaccinated with a commercially-available vaccine or given a saline injection. Then half of the pigs in each treatment group were challenged with one of two different virulent field isolates of M. hyopneumoniae and then evaluated for the ability of the vaccine to protect against disease. Pigs were evaluated for macroscopic pneumonia, local and systemic antibody production and organism level in the lungs. The dissertation concludes with a summary of General Conclusions and Future Studies. Genetic and proteomic differences between field isolates of M. hyopneumoniae impact the ability of different isolates to cause disease, and be detected and protected against. No clear traits have been definitively correlated to specific virulence factors in high versus low pathogenic isolates. This is the goal of future studies using both genetic (microarray) and proteomic (MALDI-TOF) technologies.

References 1. Ameri, M., E. M. Zhou, and W. H. Hsu. 2006. Western blot immunoassay as a confirmatory test for the presence of anti-Mycoplasma hyopneumoniae antibodies in swine serum. J. Vet. Diagn. Invest. 18:198-201. 2. Artiushin, S., and F. C. Minion. 1996. Arbitrarily primed PCR analysis of Mycoplasma hyopneumoniae field isolates demonstrates genetic heterogeneity. Int. J. Syst. Bacteriol. 46:324-328. 3. Asai, T., M. Okada, M. Ono, T. Irisawa, Y. Mori, Y. Yokomizo, and S. Sato. 1993. Increased levels of tumor necrosis factor and interleukin 1 in bronchoalveolar lavage fluids from pigs infected with Mycoplasma hyopneumoniae. Vet. Immunol. Immunopathol. 38:253-260. 4. Asai, T., M. Okada, M. Ono, Y. Mori, Y. Yokomizo, and S. Sato. 1994. Detection of interleukin-6 and prostaglandin E2 in bronchoalveolar lavage fluids of pigs 13

experimentally infected with Mycoplasma hyponeumoniae. Vet. Immunol. Immunopathol. 44:97-102. 5. Assuncao, P., N. T. Antunes, R. S. Rosales, C. Poveda, C. de la Fe, J. B. Poveda, and H. M. Davey. 2007. Application of flow cytometry for the determination of minimal inhibitory concentration of several antibacterial agents on Mycoplasma hyopneumoniae. J. Appl. Microbiol. 102:1132-1137. 6. Assuncao, P., C. De la Fe, A. S. Ramirez, O. Gonzalez Llamazares, and J. B. Poveda. 2005. Protein and antigenic variability among Mycoplasma hyopneumoniae strains by SDS-PAGE and immunoblot. Vet. Res. Commun. 29:563-74. 7. Baseman, J. B., M. Lange, N. L. Criscimagna, J. A. Giron, and C. A. Thomas. 1995. Interplay between mycoplasmas and host target cells. Microbial Pathogenesis 19:105-116. 8. Bereiter, M., T. F. Young, H. S. Joo, and R. F. Ross. 1990. Evaluation of the ELISA and comparison to the complement fixation test and radial immunodiffusion enzyme assay for detection of antibodies against Mycoplasma hyopneumoniae in swine serum. Vet. Microbiol. 25:177-192. 9. Blanchard, B., M. Kobisch, J. M. Bove, and C. Saillard. 1996. Polymerase chain reaction for Mycoplasma hyopneumoniae detection in tracheobronchiolar washings from pigs. Mol. Cell. Probes 10:15-22. 10. Blanchard, B., M. M. Vena, A. Cavalier, J. Le Lannic, J. Gouranton, and M. Kobisch. 1992. Electron microscopic observation of the respiratory tract of SPF piglets inoculated with Mycoplasma hyopneumoniae. Vet. Microbiol. 30:329-41. 11. Burnett, T. A., K. Dinkla, M. Rohde, G. S. Chhatwal, M. Srivastava, S. J. Cordwell, S. Geary, F. C. Minion, M. J. Walker, and S. P. Djordjevic. 2006. P159 is a proteolytically processed, surface adhesin of Mycoplasma hyopneumoniae: defined domains of P159 bind heparin and promote adherence to eukaryotic cells. Mol. Microbiol. 60:669-686. 12. Calsamiglia, M., and C. Pijoan. 2000. Colonisation state and colostral immunity to Mycoplasma hyopneumoniae of different parity sows. Vet. Rec. 146:530-532.

13. Calsamiglia, M., C. Pijoan, and A. Trigo. 1999. Application of a nested polymerase chain reaction assay to detect Mycoplasma hyopneumoniae from nasal swabs. J. Vet. Diagn. Invest. 11:246-251. 14. Calus, D., M. Baele, T. Meyns, A. de Kruif, P. Butaye, A. Decostere, F. Haesebrouck, and D. Maes. 2007. Protein variability among Mycoplasma hyopneumoniae isolates. Vet. Microbiol. 120:284-291. 15. Chen, J. R., J. H. Lin, C. N. Weng, and S. S. Lai. 1998. Identification of a novel adhesin-like glycoprotein from Mycoplasma hyopneumoniae. Vet. Microbiol. 62:97-110.

16. Choi, C., D. Kwon, K. Jung, Y. Ha, Y. H. Lee, O. Kim, H. K. Park, S. H. Kim, K. K. Hwang, and C. Chae. 2006. Expression of inflammatory cytokines in pigs 14

experimentally infected with Mycoplasma hyopneumoniae. J. Comp. Pathol. 134:40-46. 17. de Castro, L. A., T. R. Pedroso, S. S. Kuchiishi, M. Ramenzoni, J. D. Kich, A. Zaha, M. H. Vainstein, and H. B. Ferreira. 2006. Variable number of tandem aminoacid repeats in adhesion-related CDS products in Mycoplasma hyopneumoniae strains. Vet. Microbiol. 116:258-269. 18. DeBey, M. C., and R. F. Ross. 1994. Ciliostasis and loss of cilia induced by Mycoplasma hyopneumoniae in porcine tracheal organ cultures. Infect. Immun. 62:5312-5318. 19. Dubosson, C. R., C. Conzelmann, R. Miserez, P. Boerlin, J. Frey, W. Zimmermann, H. Hani, and P. Kuhnert. 2004. Development of two real-time PCR assays for the detection of Mycoplasma hyopneumoniae in clinical samples. Vet. Microbiol. 102:55-65. 20. Erlandson, K. R., R. B. Evans, B. J. Thacker, M. W. Wegner, and E. L. Thacker. 2005. Evaluation of three serum antibody enzyme-linked immunosorbent assays for Mycoplasma hyopneumoniae. J. Swine Health Prod. 13:198-203. 21. Fabisiak, J. P., R. D. Weiss, G. A. Powell, and J. H. Dauber. 1993. Enhanced secretion of immune-modulating cytokines by human lung fibroblasts during in vitro infection with Mycoplasma fermentans. Am. J. Respir. Cell. Mol. Biol. 8:358-364. 22. Fagan, P. K., M. J. Walker, J. Chin, G. J. Eamens, and S. P. Djordjevic. 2001. Oral immunization of swine with attenuated Salmonella typhimurium aroA SL3261 expressing a recombinant antigen of Mycoplasma hyopneumoniae (NrdF) primes the immune system for a NrdF specific secretory IgA response in the lungs. Microb. Pathog. 30:101-110. 23. Fano, E., C. Pijoan, S. Dee, and J. Deen. 2007. Effect of Mycoplasma hyopneumoniae colonization at weaning on disease severity in growing pigs. Can. J. Vet. Res. 71:195-200. 24. Friis, N. F. 1975. Some recommendations concerning primary isolation of Mycoplasma suipneumoniae and Mycoplasma flocculare a survey. Nord. Vet. Med. 27:337-339. 25. Geary, S. J., and E. M. Walczak. 1983. Cytopathic effect of whole cells and purified membranes of Mycoplasma hyopneumoniae. Infect. Immun. 41:132-136. 26. Gois, M., Cerny, M., Rozkosny, V., Sovadina, J. 1969. Studies on the epizootiological significance of some species of mycoplasma isolated from nasal swabs and lungs of pigs. Zentralbl. Veterinarmed. B. 16:253-265. 27. Goodwin, R. F. 1985. Apparent reinfection of enzootic-pneumonia-free pig herds: search for possible causes. Vet. Rec. 116:690-694. 28. Hannan, P. C. 2000. Guidelines and recommendations for antimicrobial minimum inhibitory concentration (MIC) testing against veterinary mycoplasma species. International Research Programme on Comparative Mycoplasmology. Vet. Res. 31:373-395. 15

29. Hsu, T., S. Artiushin, and F. C. Minion. 1997. Cloning and functional analysis of the P97 swine cilium adhesin gene of Mycoplasma hyopneumoniae. J. Bacteriol. 179:1317-1323. 30. Hsu, T., and F. C. Minion. 1998. Identification of the cilium binding epitope of the Mycoplasma hyopneumoniae P97 adhesin. Infect. Immun. 66:4762-4766. 31. Hsu, T., and F. C. Minion. 1998. Molecular analysis of the P97 cilium adhesin operon of Mycoplasma hyopneumoniae. Gene 214:13-23. 32. Johansson, K.-E., and B. Pettersson. 2002. Taxonomy of Mollicutes, p. 1-30. In S. Razin and R. Herrmann (ed.), Molecular Biology and Pathogenicity of Mycoplasmas. Kluwer Academic/Plenum Publishers, New York. 33. Kagnoff, M. F., and L. Eckmann. 1997. Epithelial cells as sensors for microbial infection. J. Clin. Invest. 100:6-10. 34. Keceli, S. A., and R. J. Miles. 2002. Differential inhibition of mollicute growth: an approach to development of selective media for specific mollicutes. Appl. Environ. Microbiol. 68:5012-5016. 35. King, K. W., D. H. Faulds, E. L. Rosey, and R. J. Yancey, Jr. 1997. Characterization of the gene encoding Mhp1 from Mycoplasma hyopneumoniae and examination of Mhp1's vaccine potential. Vaccine 15:25-35. 36. Kirchner, H., A. Bauer, T. Moritz, and F. Herbst. 1986. Lymphocyte activation and induction of interferon gamma in human leucocyte cultures by the mitogen in Mycoplasma arthritidis supernatant (MAS). Scand. J. Immunol. 24:609-613. 37. Kita, M., Y. Ohmoto, Y. Hirai, N. Yamaguchi, and J. Imanishi. 1992. Induction of cytokines in human peripheral blood mononuclear cells by mycoplasmas. Microbiol. Immunol. 36:507-516. 38. Kurth, K. T., T. Hsu, E. R. Snook, E. L. Thacker, B. J. Thacker, and F. C. Minion. 2002. Use of a Mycoplasma hyopneumoniae nested polymerase chain reaction test to determine the optimal sampling sites in swine. J. Vet. Diagn. Invest. 14:463- 469.

39. Lin, B. C. 2001. Intraspecies differentiation of Mycoplasma hyopneumoniae field strains isolated in the United States. Presented at the American Association of Swine Veterinarians, Kansas City, MO. 40. Madsen, M. L., D. Nettleton, E. L. Thacker, R. Edwards, and F. C. Minion. 2006. Transcriptional profiling of Mycoplasma hyopneumoniae during heat shock using microarrays. Infect. Immun. 74:160-166. 41. Madsen, M. L., M. J. Oneal, S. W. Gardner, E. L. Strait, D. Nettleton, E. L. Thacker, and F. C. Minion. 2007. Array-based genomic comparative hybridization analysis of field strains of Mycoplasma hyopneumoniae. J. Bacteriol. 189:7977-7982.

42. Maes, D., J. Segales, T. Meyns, M. Sibila, M. Pieters, and F. Haesebrouck. 2008. Control of Mycoplasma hyopneumoniae infections in pigs. Vet. Microbiol. 126:297-309. 16

43. Maniloff, J. 2002. Phylogeny and evolution, p. 31-44. In S. Razin, and R. Herrmann (ed.), Molecular biology and pathogenicity of mycoplasmas. Kluwer Academic/Plenum Publishers, New York. 44. Marois, C., J. Le Carrou, M. Kobisch, and A. V. Gautier-Bouchardon. 2007. Isolation of Mycoplasma hyopneumoniae from different sampling sites in experimentally infected and contact SPF piglets. Vet. Microbiol. 120:96-104. 45. Martelli, P., M. Terreni, S. Guazzetti, and S. Cavirani. 2006. Antibody response to Mycoplasma hyopneumoniae infection in vaccinated pigs with or without maternal antibodies induced by sow vaccination. J. Vet. Med. B. Infect. Dis. Vet. Public Health 53:229-233. 46. Mattsson, J. G., K. Bergstrom, P. Wallgren, and K. E. Johansson. 1995. Detection of Mycoplasma hyopneumoniae in nose swabs from pigs by in vitro amplification of the 16S rRNA gene. J. Clin. Microbiol. 33:893-897. 47. Mayor, D., J. Jores, B. M. Korczak, and P. Kuhnert. 2007. Multilocus sequence typing (MLST) of Mycoplasma hyopneumoniae: A diverse pathogen with limited clonality. Vet. Microbiol. 38:391-398. 48. Mayor, D., J. Jores, B. M. Korczak, and P. Kuhnert. 2008. Multilocus sequence typing (MLST) of Mycoplasma hyopneumoniae: a diverse pathogen with limited clonality. Vet. Microbiol. 127:63-72. 49. Mayor, D., F. Zeeh, J. Frey, and P. Kuhnert. 2007. Diversity of Mycoplasma hyopneumoniae in pig farms revealed by direct molecular typing of clinical material. Vet. Res. 38:391-398. 50. Messick, J. B. 2004. Hemotrophic mycoplasmas (hemoplasmas): a review and new insights into pathogenic potential. Vet. Clin. Pathol. 33:2-13. 51. Messier, S., and R. F. Ross. 1991. Interactions of Mycoplasma hyopneumoniae membranes with porcine lymphocytes. Am. J. Vet. Res. 52:1497-1502. 52. Meyns, T., D. Maes, D. Calus, S. Ribbens, J. Dewulf, K. Chiers, A. de Kruif, E. Cox, A. Decostere, and F. Haesebrouck. 2007. Interactions of highly and low virulent Mycoplasma hyopneumoniae isolates with the respiratory tract of pigs. Vet. Microbiol. 120:87-95. 53. Meyns, T., D. Maes, J. Dewulf, J. Vicca, F. Haesebrouck, and A. de Kruif. 2004. Quantification of the spread of Mycoplasma hyopneumoniae in nursery pigs using transmission experiments. Prev. Vet. Med. 66:265-275. 54. Minion, F. C., C. Adams, and T. Hsu. 2000. R1 region of P97 mediates adherence of Mycoplasma hyopneumoniae to swine cilia. Infect. Immun. 68:3056-3060. 55. Minion, F. C., E. J. Lefkowitz, M. L. Madsen, B. J. Cleary, S. M. Swartzell, and G. G. Mahairas. 2004. The genome sequence of Mycoplasma hyopneumoniae strain 232, the agent of swine mycoplasmosis. J. Bacteriol. 186:7123-7133.

56. Muneta, Y., Y. Minagawa, Y. Shimoji, R. Nagata, P. F. Markham, G. F. Browning, and Y. Mori. 2006. IL-18 expression in pigs following infection with Mycoplasma hyopneumoniae. J. Interferon. Cytokine. Res. 26:637-644. 17

57. Muneta, Y., H. Uenishi, R. Kikuma, K. Yoshihara, Y. Shimoji, R. Yamamoto, N. Hamashima, Y. Yokomizo, and Y. Mori. 2003. Porcine TLR2 and TLR6: identification and their involvement in Mycoplasma hyopneumoniae infection. J. Interferon Cytokine Res. 23:583-90. 58. Muto, A., and C. Ushida. 2002. Transcription and translation.p. 323-345. In S. R. a. R. Herrmann (ed.), Molecular biology and pathogenicity of mycoplasmas. Plenum Press, New York. 59. Neimark, H., K. E. Johansson, Y. Rikihisa, and J. G. Tully. 2001. Proposal to transfer some members of the genera Haemobartonella and Eperythrozoon to the genus Mycoplasma with descriptions of 'Candidatus Mycoplasma haemofelis', 'Candidatus Mycoplasma haemomuris', 'Candidatus Mycoplasma haemosuis' and 'Candidatus Mycoplasma wenyonii'. Int. J. Syst. Evol Microbiol. 51:891-899. 60. Neimark, H., K. E. Johansson, Y. Rikihisa, and J. G. Tully. 2002. Revision of haemotrophic Mycoplasma species names. Int. J. Syst. Evol. Microbiol. 52:683. 61. Nicolet, J., P. Paroz, and S. Bruggmann. 1980. Tween 20 soluble proteins of Mycoplasma hyopneumoniae as antigen for an enzyme linked immunosorbent assay. Res. Vet. Sci. 29:305-309. 62. Opriessnig, T., E. L. Thacker, S. Yu, M. Fenaux, X. J. Meng, and P. G. Halbur. 2004. Experimental reproduction of postweaning multisystemic wasting syndrome in pigs by dual infection with Mycoplasma hyopneumoniae and porcine circovirus type 2. Vet. Pathol. 41:624-640. 63. Ossewaarde, J. M., A. de Vries, T. Bestebroer, and A. F. Angulo. 1996. Application of a mycoplasma group-specific PCR for monitoring decontamination of mycoplasma-infected Chlamydia sp. strains. Appl. Environ. Microbiol. 62:328- 331. 64. Park, S. C., S. Yibchok-Anun, H. Cheng, T. F. Young, E. L. Thacker, F. C. Minion, R. F. Ross, and W. H. Hsu. 2002. Mycoplasma hyopneumoniae increases intracellular calcium release in porcine ciliated tracheal cells. Infect. Immun. 70:2502-2506. 65. Piffer, I. A., and R. F. Ross. 1984. Effect of age on susceptibility of pigs to Mycoplasma hyopneumoniae pneumonia. Am. J. Vet. Res. 45:478-481. 66. Piffer, I. A., and R. F. Ross. 1985. Immunofluorescence technique for detection of Mycoplasma hyopneumoniae in swine lungs. Pesq. agropec. bras. 20:877-882. 67. Potgieter, L. N., M. L. Frey, and R. F. Ross. 1972. Chronological development of Mycoplasma hyorhinis and Mycoplasma hyosynoviae infections in cultures of a swine synovial cell strain. Can. J. Comp. Med. 36:145-149. 68. Potgieter, L. N., and R. F. Ross. 1972. Demonostration of Mycoplasma hyorhinis and Mycoplasma hyosynoviae in lesions of experimentally infected swine by immunofluorescence. Am. J. Vet. Res. 33:99-105.

69. Razin, S. 1978. The mycoplasmas. Microbiol. Rev. 42:414-470. 18

70. Razin, S., and E. A. Freundt. 1984. The mycoplasmas, p. 740-793. In N. R. K. a. J. G. Holt (ed.), Bergey's manual of systemic bacteriology, vol. 1. Williams and Wilkins, Baltimore. 71. Razin, S., D. Yogev, and Y. Naot. 1998. Molecular biology and pathogenicity of mycoplasmas. Microbiol. Mol. Biol. Rev. 62:1094-1156. 72. Ro, L. H., and R. F. Ross. 1983. Comparison of Mycoplasma hyopneumoniae strains by serologic methods. Am. J. Vet. Res. 44:2087-2094. 73. Roberts, D. H. 1974. Experimental infection of pigs with Mycoplasma hyopneumoniae (suipneumoniae). Br. Vet. J. 130:68-74. 74. Rodriguez, F., O. Quesada, J. B. Poveda, A. Fernandez, and H. Lorenzo. 2007. Immunohistochemical detection of interleukin-12 and interferon-gamma in pigs experimentally infected with Mycoplasma hyopneumoniae. J. Comp. Pathol. 136:79-82. 75. Rodriguez, F., G. A. Ramirez, J. Sarradell, M. Andrada, and H. Lorenzo. 2004. Immunohistochemical labelling of cytokines in lung lesions of pigs naturally infected with Mycoplasma hyopneumoniae. J. Comp. Pathol. 130:306-312. 76. Ross, R. F., B. J. Zimmermann-Erickson, and T. F. Young. 1984. Characteristics of protective activity of Mycoplasma hyopneumoniae vaccine. Am. J. Vet. Res. 45:1899-1905. 77. Rottem, S. 2002. Invasion of mycoplasmas into and fusion with host cells, p. 391- 401. In S. Razin, and R. Herrmann (ed.), Molecular biology and pathogenicity of mycoplasmas. Kluwer Academic/Plenum Publishers, New York. 78. Sarradell, J., M. Andrada, A. S. Ramirez, A. Fernandez, J. C. Gomez- Villamandos, A. Jover, H. Lorenzo, P. Herraez, and F. Rodriguez. 2003. A morphologic and immunohistochemical study of the bronchus-associated lymphoid tissue of pigs naturally infected with Mycoplasma hyopneumoniae. Vet. Pathol. 40:395-404. 79. Scarman, A. L., J. C. Chin, G. J. Eamens, S. F. Delaney, and S. P. Djordjevic. 1997. Identification of novel species-specific antigens of Mycoplasma hyopneumoniae by preparative SDS-PAGE ELISA profiling. Microbiology 143:663-673. 80. Shimoji, Y., E. Oishi, Y. Muneta, H. Nosaka, and Y. Mori. 2003. Vaccine efficacy of the attenuated Erysipelothrix rhusiopathiae YS-19 expressing a recombinant protein of Mycoplasma hyopneumoniae P97 adhesin against mycoplasmal pneumonia of swine. Vaccine 21:532-537. 81. Sibila, M., M. Calsamiglia, D. Vidal, L. Badiella, A. Aldaz, and J. C. Jensen. 2004. Dynamics of Mycoplasma hyopneumoniae infection in 12 farms with different production systems. Can. J. Vet. Res. 68:12-18. 82. Sibila, M., M. Nofrarias, S. Lopez-Soria, J. Segales, P. Riera, D. Llopart, and M. Calsamiglia. 2007. Exploratory field study on Mycoplasma hyopneumoniae infection in suckling pigs. Vet. Microbiol. 121:352-356. 19

83. Simecka, J. W., J. K. Davis, M. K. Davidson, S. E. Ross, C. T. K.-H. Stadtlander, and G. H. Cassell. 1992. Mycoplasma diseases of animals, p. 391-415. In J. Maniloff, R. N. McElhaney, L. R. Finch, and J. B. Baseman (ed.), Mycoplasmas: molecular biology and pathogenesis. American Society for Microbiology, Washington, D. C. 84. Smith, A. 1992. Eperythrozoonosis, p. 470-474. In B. Straw, Mengeling, W.L., D'Allaire, S., Taylor, D. J. (ed.), Diseases of Swine. Iowa State University Press, Ames. 85. Sorensen, V., P. Ahrens, K. Barfod, A. A. Feenstra, N. C. Feld, N. F. Friis, V. Bille-Hansen, N. E. Jensen, and M. W. Pedersen. 1997. Mycoplasma hyopneumoniae infection in pigs: duration of the disease and evaluation of four diagnostic assays. Vet. Microbiol. 54:23-34. 86. Stakenborg, T., J. Vicca, P. Butaye, H. Imberechts, J. Peeters, A. De Kruif, F. Haesebrouck, and D. Maes. 2006. A multiplex PCR to identify porcine mycoplasmas present in broth cultures. Vet. Res. Commun. 30:239-247. 87. Stakenborg, T., J. Vicca, D. Maes, J. Peeters, A. de Kruif, F. Haesebrouck, and P. Butaye. 2006. Comparison of molecular techniques for the typing of Mycoplasma hyopneumoniae isolates. J. Microbiol. Methods 66:263-275. 88. Stark, K. D. 2000. Epidemiological investigation of the influence of environmental risk factors on respiratory diseases in swine-A literature review. Vet. J. 159:37-56. 89. Stark, K. D., J. Nicolet, and J. Frey. 1998. Detection of Mycoplasma hyopneumoniae by air sampling with a nested PCR assay. Appl. Environ. Microbiol. 64:543-548. 90. Stemke, G. W. 1997. Gene amplification (PCR) to detect and differentiate mycoplasmas in porcine mycoplasmal pneumonia. Lett. Appl. Microbiol. 25:327- 330. 91. Stemke, G. W., and J. A. Robertson. 1982. Comparison of two methods for enumeration of mycoplasmas. J. Clin. Microbiol. 16:959-961.

92. Strait, E., B. Erickson, and E. Thacker. 2004. Analysis of Mycoplasma hyopneumoniae field isolates. Presented at the Proc AASV, Des Moines, IA. 93. Tajima, M., and T. Yagihashi. 1982. Interaction of Mycoplasma hyopneumoniae with the porcine respiratory epithelium as observed by electron microscopy. Infect. Immun. 37:1162-1169. 94. Tajima, M., T. Yagihashi, T. Nunoya, A. Takeuchi, and F. Ohashi. 1984. Mycoplasma hyopneumoniae infection in pigs immunosuppressed by thymectomy and treatment with antithymocyte serum. Am. J. Vet. Res. 45:1928-32. 95. Tang, J., M. Hu, S. Lee, and R. Roblin. 2000. A polymerase chain reaction based method for detecting mycoplasma/acholeplasma contaminants in cell culture. J. Microbiol. Methods 39:121-126. 20

96. Thacker, E. L., P. G. Halbur, R. F. Ross, R. Thanawongnuwech, and B. J. Thacker. 1999. Mycoplasma hyopneumoniae potentiation of porcine reproductive and respiratory syndrome virus-induced pneumonia. J. Clin. Microbiol. 37:620- 627. 97. Thanawongnuwech, R., and E. L. Thacker. 2003. Interleukin-10, interleukin-12, and interferon-gamma levels in the respiratory tract following Mycoplasma hyopneumoniae and PRRSV infection in pigs. Viral. Immunol. 16:357-367. 98. Uphoff, C. C., and H. G. Drexler. 2002. Comparative PCR analysis for detection of mycoplasma infections in continuous cell lines. In Vitro Cell Dev. Biol. Anim. 38:79-85. 99. Vasconcelos, A. T., H. B. Ferreira, C. V. Bizarro, S. L. Bonatto, M. O. Carvalho, P. M. Pinto, D. F. Almeida, L. G. Almeida, R. Almeida, L. Alves-Filho, E. N. Assuncao, V. A. Azevedo, M. R. Bogo, M. M. Brigido, M. Brocchi, H. A. Burity, A. A. Camargo, S. S. Camargo, M. S. Carepo, D. M. Carraro, J. C. de Mattos Cascardo, L. A. Castro, G. Cavalcanti, G. Chemale, R. G. Collevatti, C. W. Cunha, B. Dallagiovanna, B. P. Dambros, O. A. Dellagostin, C. Falcao, F. Fantinatti-Garboggini, M. S. Felipe, L. Fiorentin, G. R. Franco, N. S. Freitas, D. Frias, T. B. Grangeiro, E. C. Grisard, C. T. Guimaraes, M. Hungria, S. N. Jardim, M. A. Krieger, J. P. Laurino, L. F. Lima, M. I. Lopes, E. L. Loreto, H. M. Madeira, G. P. Manfio, A. Q. Maranhao, C. T. Martinkovics, S. R. Medeiros, M. A. Moreira, M. Neiva, C. E. Ramalho-Neto, M. F. Nicolas, S. C. Oliveira, R. F. Paixao, F. O. Pedrosa, S. D. Pena, M. Pereira, L. Pereira-Ferrari, I. Piffer, L. S. Pinto, D. P. Potrich, A. C. Salim, F. R. Santos, R. Schmitt, M. P. Schneider, A. Schrank, I. S. Schrank, A. F. Schuck, H. N. Seuanez, D. W. Silva, R. Silva, S. C. Silva, C. M. Soares, K. R. Souza, R. C. Souza, C. C. Staats, M. B. Steffens, S. M. Teixeira, T. P. Urmenyi, M. H. Vainstein, L. W. Zuccherato, A. J. Simpson, and A. Zaha. 2005. Swine and poultry pathogens: the complete genome sequences of two strains of Mycoplasma hyopneumoniae and a strain of Mycoplasma synoviae. J. Bacteriol. 187:5568-5577. 100. Verdin, E., C. Saillard, A. Labbe, J. M. Bove, and M. Kobisch. 2000. A nested PCR assay for the detection of Mycoplasma hyopneumoniae in tracheobronchiolar washings from pigs. Vet. Microbiol. 76:31-40. 101. Vicca, J., D. Maes, L. Thermote, J. Peeters, F. Haesebrouck, and A. de Kruif. 2002. Patterns of Mycoplasma hyopneumoniae infections in Belgian farrow-to- finish pig herds with diverging disease-course. J. Vet. Med. B. Infect. Dis. Vet. Public Health 49:349-353. 102. Vicca, J., T. Stakenborg, D. Maes, P. Butaye, J. Peeters, A. de Kruif, and F. Haesebrouck. 2003. Evaluation of virulence of Mycoplasma hyopneumoniae field isolates. Vet. Microbiol. 97:177-190. 103. Wallgren, P., G. Bolske, S. Gustafsson, S. Mattsson, and C. Fossum. 1998. Humoral immune responses to Mycoplasma hyopneumoniae in sows and offspring following an outbreak of mycoplasmosis. Vet. Microbiol. 60:193-205. 104. Yamao, F., A. Muto, Y. Kawauchi, M. Iwami, S. Iwagami, Y. Azumi, and S. Osawa. 1985. UGA is read as tryptophan in Mycoplasma capricolum. Proc. Natl. Acac. Sci. USA 82:2306-2309. 21

105. Zhang, Q., T. F. Young, and R. F. Ross. 1995. Identification and characterization of a Mycoplasma hyopneumoniae adhesin. Infect. Immun. 63:1013-1019. 106. Zhao, S. H., D. Kuhar, J. K. Lunney, H. Dawson, C. Guidry, J. J. Uthe, S. M. Bearson, J. Recknor, D. Nettleton, and C. K. Tuggle. 2006. Gene expression profiling in Salmonella choleraesuis-infected porcine lung using a long oligonucleotide microarray. Mamm. Genome 17:777-89. 107. Zielinski, G. C., and R. F. Ross. 1993. Adherence of Mycoplasma hyopneumoniae to porcine ciliated respiratory tract cells. Am. J. Vet. Res. 54:1262-1269.

CHAPTER 2. COMPARISON OF MYCOPLASMA HYOPNEUMONIAE FIELD ISOLATES

E. L. Strait, B. Z. Erickson, F. C. Minion, E. L. Thacker

Dept. of Veterinary Microbiology and Preventive Medicine, College of Veterinary Medicine, Iowa State University, Ames, IA

In preparation for submission to Veterinary Microbiology

Abstract In vivo and in vitro characteristics associated with four field isolates of Mycoplasma hyopneumoniae and strain 232 were compared in a challenge study. Pigs were intratracheally inoculated on three consecutive days with log-phase culture of one of the isolates, or on one day with lung homogenate containing strain 232. Virulence was measured as the ability to induce macroscopic lung lesions. Virulence was then correlated to immunoglobulin (Ig)A and IgG levels in bronchial alveolar lavage fluid (BALF), interleukin (IL) 1ß and IL-18 cytokine levels in BALF, and to in vitro and in vivo growth characteristics. In addition, serum antibody levels were tested using three different M. hyopneumoniae-specific ELISAs: the Tween 20 ELISA, the HerdChek® Mycoplasma hyopneumoniae ELISA (IDEXX Laboratories, Westbrook, Maine), and the IDEIA™ M. hyopneumoniae EIA KIT (Oxoid, Ely, Cambridgeshire) to compare the sensitivity of these assays. In this study significant differences in the amount of macroscopic pneumonia were produced between the challenge groups, but no trait was identified that significantly correlated to the level of virulence observed in the isolates. The Oxoid ELISA was the most sensitive assay for detecting antibodies in serum tested at 28 days-post-challenge (DPC). The Tween 20 and IDEXX ELISAs detected fewer positive results at 28 DPC and had discrepant results between the two assays in 34% of these samples. This study confirmed variations in virulence among field isolates and suggests differences in the ability of ELISA assays to detect antibodies raised isolates of M. hyopneumoniae in the field.

1. Introduction Mycoplasma hyopneumoniae is widely considered to be one of the most important pathogens affecting pigs today (Holtkamp, 2007). As a primary pathogen, M. hyopneumoniae produces a bronchopneumonia characterized by peribronchiolar and perivascular lymphoid 23

hyperplasia with high morbidity and low mortality. More importantly, M. hyopneumoniae combines with other bacteria and viruses to form the porcine respiratory disease complex. A previous study by Vicca, et al. (2003) identified field isolates of M. hyopneumoniae with high, moderate or low virulence, but traits correlating with virulence remain poorly defined. Vicca, et al. (2003) identified a 5,000 base pair fragment by randomly amplified polymorphic DNA (RAPD) analysis that appeared to be associated with more virulent strains, but no function has yet been assigned to this fragment. Using one high virulent and one low virulent isolate from the Vicca collection (Vicca et al., 2003) in a challenge study using cesarean- derived, colostrum-deprived (CD/CD) pigs, Meyns, et al. (2007) reported increased virulence associated with faster in vitro growth, an enhanced capacity to multiply in the lungs, the ability to induce increased inflammatory cytokines interleukin (IL)-1ß and TNFα and attract increased numbers of neutrophils in the lungs. These associations were made using only two isolates; therefore, more isolates need to be studied to determine if these correlations apply in general. This study sought to compare virulence among isolates of M. hyopneumoniae by experimentally challenging pigs using culture of four recent field isolates, or a well- established model using lung homogenate from strain 232 which derives from M. hyopneumoniae strain 11 (Ross et al., 1984; Thacker et al., 1999; Thacker et al., 2000a). The four M. hyopneumoniae field isolates used in this study have been previously shown to be genetically distinct from each other and from strain 232 (Madsen et al., 2007). Necropsies were performed at 28 days-post-challenge (DPC) and virulence was assessed by comparing macroscopic lung lesions. The inflammatory cytokines IL-1ß and IL-18, previously shown to be increased during M. hyopneumoniae infection were measured in bronchial alveolar lavage fluid (BALF) (Asai et al., 1993; Lorenzo et al., 2006; Meyns et al., 2007; Muneta et al., 2006). In addition from BALF, immunoglobulin (Ig)A and IgG levels were measured by ELISA (Thacker et al., 2000b), and organism level was determined by real-time PCR. Serum collected prior to challenge and at necropsy 28 DPC was tested using three different M. hyopneumoniae-specific ELISAs: the Tween 20 ELISA(Bereiter et al., 1990), the HerdChek® Mycoplasma hyopneumoniae ELISA (IDEXX Laboratories, Westbrook, Maine), and the IDEIA™ M. hyopneumoniae EIA KIT (Oxoid, Ely, Cambridgeshire) to compare the sensitivity of these assays. In addition, in vitro growth characteristics were compared using real-time PCR to generate growth curves. The in vitro growth rates and capacity to stimulate an immune response measured by local cytokine and immunoglobulin levels were correlated to 24

macroscopic pneumonia in order to identify traits related to increased virulence in field isolates of M. hyopneumoniae.

2. Materials and Methods

2.1. Mycoplasma hyopneumoniae field isolates Mycoplasma hyopneumoniae isolates: 95MP1505, 95MP1505, 00MP1301, and 00MP1502 were isolated from pigs presented to the Iowa State University Veterinary Diagnostic Laboratory for necropsy. Tissue was aseptically collected at the leading edge of lesions from lungs of pigs with pneumonia characteristic of M. hyopneumoniae. The lung tissue was ground in Friis media (Friis, 1975) and inoculated by 10-fold dilution into a series of culture tubes with Friis media. Cultures indicating growth by turbidity and acid change in the media were confirmed as M. hyopneumoniae using a previously published multiplex PCR assay (Stakenborg et al., 2006).

2.2. Study design Thirty-eight 10-to 12-day-old pigs from a farm serologically negative for M. hyopneumoniae, porcine reproductive and respiratory syndrome virus and swine influenza virus were divided into five challenge groups of seven pigs and one negative control group of three pigs. Groups were individually housed in separate rooms in a biosecure animal holding facility at Iowa State University and fed ad libitum throughout the study. The four groups challenged with the field isolates were inoculated on three consecutive days (0 DPC, 1 DPC and 2 DPC) with 10 ml of log-phase culture media as determined by turbidity and an acidification of the media to approximately pH 6.8 (Zielinski and Ross, 1990). Cultures for isolates 00MP1301 and 00MP1502 were at passage six on challenge days one and two and passage seven for challenge day three. Cultures for isolates 95MP1505 and 95MP1509 were at passage eight on challenge days one and two and passage nine for challenge day three. The group receiving M. hyopneumoniae strain 232 was inoculated on 0 DPC with 10 ml of diluted lung homogenate as previously described (Ross et al., 1984). Color-changing unit (CCU) titers for each inoculum were determined each challenge day as previously described (Stemke and Robertson, 1982). The negative control group received no treatment. Pigs in all groups were necropsied at 28 DPC. All animal procedures and care were conducted under the supervision and regulations of the Iowa State University Institutional Animal Care and Use Committee. 25

2.3. Sample collection Blood was collected for serum at 0 DPC and at necropsy, 28 DPC. Lesions consistent with mycoplasmal pneumonia were sketched on a standard lung diagram then analyzed for the proportion of affected lung using a Zeiss SEM-IPS image analysis system (Ross et al., 1984). Bronchial alveolar lavage fluid (BALF) was collected from each pig using 50 ml of sterile phosphate-buffered saline (PBS), pH 7.2 (1.54mM KH2PO4, 155.17mM NaCl, and

2.71mM Na2HPO4-7H20) as previously described (Mengeling et al., 1995). After collection, the BALF was divided into 1 ml aliquots and frozen at -80ºC until assayed by quantitative real-time PCR and cytokine ELISAs as described below. Bronchial swabs were collected for mycoplasma isolation and also for routine bacterial culture.

2.4. Serology Three ELISA’s for M. hyopneumoniae antibody detection: the Tween 20 (Bereiter et al., 1990), the HerdChek® Mycoplasma hyopneumoniae ELISA, and the IDEIA™ M. hyopneumoniae EIA KIT were used to measure antibodies to M. hyopneumoniae at 0 and 28 DPC. The Tween 20 was performed as previously described (Bereiter et al., 1990) and the IDEXX and Oxoid ELISAs were performed as per manufacturer’s instructions. In the Oxoid ELISA, a sample is defined as positive if the percent inhibition of the sample is ≤ 50% of the buffer control. Samples with an OD > 50% of the buffer control are defined as negative. The IDEXX assay presents results as a S:P ratio which is calculated as: (sample OD – negative control OD) ÷ (positive control OD – negative control OD). A S:P ratio ≥ 0.4 is positive; < 0.4, ≥ 0.3 is suspect; and < 0.3 is negative. For the Tween 20 ELISA, all samples are tested in duplicate and the average optical density (OD) of the two is used. The positive control is normalized to an OD of 0.4, and the samples are then adjusted accordingly. Optical densities ≥ 0.24 are defined as positive. Negative results are defined as OD < 0.2, and an OD < 0.24, ≥ 0.2 is defined as suspect.

2.5. BALF analysis To quantify the number of Mycoplasma hyopneumoniae organisms in the lungs, a real- time PCR was performed on a 1 mL aliquot of BALF from each pig. The BALF was centrifuged at 16,000 xg for 10 minutes, and the resulting pellet was resuspended in 200 µl of PBS. DNA was isolated using the QIAamp DNA mini kit (Qiagen, Valencia, California) as per manufacturer’s instructions. In addition, DNA was isolated from M. hyopneumoniae strain 232 by phenol-chloroform extraction (Sambrook and David, 2001) for use in generating a standard curve for quantitative analysis. A real-time PCR assay targeting the 26

mhp183 gene of M. hyopneumoniae was performed as previously described (Strait et al., submitted). Commercially available immunoassays for IL-1ß and IL-18 (BioSource™ Swine Immunoassay Kit, Invitrogen, Carlsbad, CA) were performed on BALF as per manufacturer’s instructions including quantification by standard curve. M. hyopneumoniae- specific IgA and IgG antibodies were measured using a previously described assay (Thacker et al., 2000b).

2.6. Growth curves A quantitative real-time PCR targeting the gene mhp183 (Strait, et al., submitted) was used to generate a growth-curve for each M. hyopneumoniae isolate. Tubes containing 5 ml of Friis media were inoculated with 500 µl of log-phase culture from each isolate. For the growth-curve, 200 µl samples were collected at multiple time-points over a 96 hr period. The DNA from each culture was extracted as described above using the QIAamp DNA mini kit, and then assayed using the mhp183 real-time PCR. A standard curve was generated using phenol-chloroform extracted DNA from strain 232 (Sambrook and David, 2001).

2.7. Statistics Statistical differences between groups were analyzed using Analysis of Variance (ANOVA) on data including percentage of lung lesions, IgA and IgG ELISA antibody levels, IL-1ß and IL-18 ELISA cytokine levels, and organism levels measured by quantitative real- time PCR. Least significant differences (LSD) were calculated to compare means. All analyses were considered significant at p < 0.05.

3. Results

3.1. Challenge inoculum titers Titration of each inoculum was performed on each day of challenge. The inoculum used for the four field isolates were log-phase cultures and the challenge procedure was performed on three consecutive days. Their average titers were as follows: 107.3 CCU/ml for strain 95MP1505; 107.2 CCU/ml for strain 95MP1509; and 104.2 CCU/ml for strains 00MP1301 and 00MP1502. Pigs challenged with strain 232 were given diluted lung inoculum once at a titer of 104 CCU/ml.

3.2. Percent pneumonia All pigs in the M. hyopneumoniae challenge groups produced lesions characteristic of mycoplasmal pneumonia, while the negative controls had no visible signs of pneumonia (Figure 2.1 and Table 2.1). The percentage of pneumonia caused by isolate 95MP1509 was significantly higher than was caused by isolates, 00MP1502 or strain 232. The percentage of pneumonia caused by isolate 00MP1301 was significantly greater than observed with isolate 00MP1502.

Figure 2.1. Box-plot comparing the percent of macroscopic pneumonia in pigs challenged with one of four field isolates of M. hyopneumoniae, strain 232, or unchallenged controls. Groups indicated by different lower-case letters are significantly different from each other, p < 0.05. *Denotes an outlier that was excluded from the statistical analysis.

3.3. Bacterial isolation Mycoplasma hyopneumoniae was isolated from bronchial swabs collected at necropsy from all pigs in the groups challenged with isolates 95MP1505, 00MP1502, and strain 232. In the group challenged with isolate 00MP1301, 6/7 pigs cultured positive for M. hyopneumoniae, while 4/7 pigs cultured positive in the 95MP1509 group. Streptococcus suis was cultured from 2 pigs in each of the 00MP1301 and 95MP1509 groups, but no disease was found to be associated with these infections.

Table 2.1. Summary of pneumonia and bronchial alveolar lavage fluid analysis at 28 DPC. PIG Pneumonia IL-1B IL-18 IgA IgG GROUP ID (%) PCR (ng/ul) (pg/ml) (pg/ml) (OD) (OD) 410 1.19 0.004 0 0 0.18 0.004 411 2.72 0.193 10.37 0 0.432 0.011 420 1.15 0.032 3.44 0.11 0.044 0 421 0.44 0.148 0 0 0.205 0.037 429 8.57 2.012 0.12 0 0.287 0.105 432 11.01 3.664 7.59 23.39 0.331 0.076 232 438 14.22 0.926 24.55 0 0.347 0.192 407 10.58 1.005 17.08 0 0.464 0.126 412 14.08 2.301 1.71 0 0.638 0.195 417 5.22 1.152 0 0 0.611 0.146 418 7.79 1.743 0 0 0.57 0.18 427 1.21 1.372 0 0 0.39 0.023 431 17.47 3.727 9.83 0.55 0.529 0.488 00MP1301 437 20.30 2.860 3.17 0 0.424 0.262 409 4.96 3.619 0 0 0.403 0.106 413 4.83 1.623 9.92 4.50 0.36 0.039 414 7.32 2.866 0 0 0.446 0.105 416 6.74 5.263 3.49 0 0.397 0.27 430 10.20 1.427 6.68 0 0.192 0.332 433 7.39 1.292 110.67 94.11 0.383 0.132 95MP1505 434 24.42 2.559 45.48 110.02 0.552 0.627 406 6.61 1.623292 7.27 0 0.633 0.092 415 10.11 0.71398 21.59 0 0.571 0.216 422 5.92 0.890401 9.42 13.35 0.691 0.211 423 10.96 0.827501 0.89 0 0.637 0.115 425 9.87 0.443709 0 0 0.673 0.208 436 29.88 5.060982 2.99 0 0.467 0.372 95MP1509 439 21.13 4.664796 26.60 155.21 0.566 0.207 404 0.75 1.163191 32.85 21.30 0.438 0.015 405 2.54 0.392438 7.78 12.20 0.695 0.304 419 2.08 1.068819 2.76 0 0.377 0.037 424 2.06 4.434803 0.85 0 0.577 0.21 426 6.73 5.060982 38.14 37.21 0.363 0.169 428 34.20 10.627708 42.512 16.797 0.372 0.387 00MP1502 435 1.48 2.612066 67.54 37.68 0.501 0.233 402 0.00 0 0 0 0 0 403 0.00 0 0.755 0 0.048 0 control 408 0.00 0 6.818 0 0 0 ID = individual pig identification; OD = optical density. 29

3.4. Quantitative real-time PCR Real-time PCR was used to quantify the number of M. hyopneumoniae organisms in the lungs of pigs at necropsy by BALF analysis. A summary of these results is in Table 2.1. No significant difference in the number of organisms in BALF was found between any of the challenge groups.

3.5. IL-1ß and IL-18 ELISAs Quantitative cytokine ELISAs measuring IL-1ß and IL-18 in BALF were performed. Results from these assays are presented in Table 2.1. No significant difference in either cytokine was found between any of the groups, including the negative controls.

3.6. IgA and IgG ELISAs Immunoglobulin A and IgG ELISAs were performed on BALF from each pig (Table 2.1). No significant difference in either immunoglobulin was found between any of the challenge groups, but all had significantly more IgA antibodies than the negative control groups (p<0.0001). For IgG antibodies, the difference between the challenged and unchallenged groups approached significance (p=0.0538).

3.7. Serology All pigs were seronegative prior to challenge and the control group remained negative throughout the trial as measured by all three ELISA’s. At 28 DPI, the Oxoid ELISA found all pigs positive except for one pig that was negative. Overall, the IDEXX assay detected 40% of the samples as positive, 43% as negative and 17% as suspect, while the Tween 20 assay detected 46% samples as positive, 46% as negative and 8% as suspect. A direct comparison between the Tween 20 and IDEXX ELISAs showed that 12/35 of the results from individual pigs were discrepant between these two tests (Table 2.2).

3.8. Growth curves A comparison of growth curves for each of the four field isolates and strain 232 is shown in Figure 2.2. Isolates 95MP1505, 00MP1502 and 00MP1301 grew to the highest titer while isolate 95MP1509 and strain 232 had the lowest titers. Isolates 95MP1505 and 95MP1509 reached their maximum titer more quickly than did the other cultures.

Table 2.2. A comparison of Tween 20, IDEXX and Oxoid ELISA results from sera collected at necropsy (28 DPC) from pigs challenged with one of four field isolates of M. hyopneumoniae or strain 232.

Challenge group ELISA Pig replicates (7/group) 232 Tween 20 - - - - +/- + + IDEXX ------+ Oxoid + + - + + + + 00MP1502 Tween 20 - - - - - + + IDEXX - - - - +/- +/- + Oxoid + + + + + + + 00MP1301 Tween 20 - - +/- +/- + + + IDEXX - +/- +/- + - - + Oxoid + + + + + + + 95MP1505 Tween 20 - - - + + + + IDEXX - - +/- + + + + Oxoid + + + + + + + 95MP1509 Tween 20 - - + + + + + IDEXX + + +/- + + + + Oxoid + + + + + + +

Growth curve comparison

7 6 00MP1301 5 (ng/ul) 00MP1502 4 95MP1505 3 95MP1509 2 232 1 Concentration 0 0 15 23 39 66 70 88 Time (hr)

Figure 2.2. Growth-curves for four M. hyopneumoniae field isolates and strain 232 generated from real-time PCR on culture samples.

4. Discussion The results of this study confirm previous findings demonstrating a range of virulence between field isolates of M. hyopneumoniae measured at 28 DPC. In the current study no single measure of the immune response in pigs, or growth characteristics of the organism in vivo or in vitro was predictive for level of virulence in the field isolates. This is in contrast to the study by Meyns, et al. (2007) that found a correlation of increased virulence to faster in vitro growth, increased organism level and numbers of neutrophils in the lungs, and the ability to induce the inflammatory cytokines IL-1ß and TNFα (Meyns et al., 2007). The study by Meyns et al. (2007) compared a highly virulent isolate of M. hyopneumoniae to an isolate cultured from a herd with no clinical signs using CD/CD pigs. In the current study conventional pigs were used and the isolates for challenge were obtained from clinical cases submitted for necropsy at the Iowa State University Veterinary Diagnostic Laboratory. Therefore, the difference in virulence between the groups was less pronounced than in the studies comparing field isolates of M. hyopneumoniae cultured from herds with varying disease profiles (Meyns et al., 2007; Vicca et al., 2003). As is common in M. hyopneumoniae challenge studies, considerable pig-to-pig variation was observed within each group and statistical significance when comparing lung lesions was achieved only after an outlier from group 00MP1502 was removed from the analysis. Similarly, the combination of small group numbers and individual variation impeded statistical analysis of IL-1ß and IL-18 cytokine levels, and IgA and IgG antibody levels in BALF. It is unclear why IL-18 cytokine levels were not detected in over half of the BALF samples as this cytokine has been demonstrated at high levels by ELISA and immunohistochemistry at 28 DPC in a previous study (Muneta et al., 2006). Perhaps the ELISA used in the current study was less sensitive, or peak levels of secretion occurred at an earlier time-point. Mycoplasma hyopneumoniae forms small, difficult to count colonies on agar; therefore, quantification using standard bacterial culture methods is not possible. Instead, a quantitative real-time PCR targeting the mhp183 gene of M. hyopneumoniae was used to generate a growth-curve for each M. hyopneumoniae isolate in this study. The growth curves and quantitative real-time PCR assay measuring organism level in the lung in the current study failed to support the findings by Meyns, et al. (2007) that correlated enhanced growth characteristics in vitro and in vivo measured using the CCU method with increased virulence. Additionally, differences in CCU titers of challenge inoculum in this study did not correlate to lung lesions at necropsy. Whereas inoculum for isolates 95MP1505 and 95MP1509 had CCU/ml titers 1000-fold higher than isolates 00MP1301 and 00MP1502 and strain 232, 32

isolates 95MP1509 and 00MP1301 induced the highest amount of lung lesions and 95MP1505, 00MP1502 and strain 232 induced the lowest percentage of lesions. The three ELISA’s for M. hyopneumoniae antibody detection tested in this study differed in their sensitivity for antibodies produced against the different M. hyopneumoniae isolates. Previous comparisons of these ELISAs have relied on field sera or sera from pigs challenged only with strain 232 (Ameri-Mahabadi et al., 2005; Erlandson et al., 2005). No other experiments have previously compared sera of known status from pigs infected with different isolates of M. hyopneumoniae. At 28 DPC every challenged pig except one was positive by the Oxoid™ ELISA. While the IDEXX and Tween 20 ELISAs produced similar numbers of positive, negative and suspect results overall when compared to each other, each produced fewer positives from challenged pigs at 28 DPI compared to the Oxoid ELISA as has been previously demonstrated (Erlandson et al., 2005). Although when the IDEXX and Tween 20 ELISAs were compared on an individual pig basis they disagreed 34% of the time. Both the IDEXX and Tween 20 ELISAs are produced based on the same methodology (Bereiter et al., 1990) with the strain of M. hyopneumoniae used for antigen production as the primary difference between the two assays. Therefore, the degree of individual variation in results between these two assays is somewhat unexpected. The differences between the two indirect ELISAs leading to the discrepant results was not identified in this study. The results of this challenge study confirm that differences in virulence exist among field isolates of M. hyopneumoniae compared at 28 DPC. More studies using a wide range of field isolates using larger challenge groups are required to identify traits that correlate with level of virulence. The impact of antigenic differences between field isolates of M. hyopneumoniae on serology results also warrants further investigation.

Acknowledgements This project was funded by a grant from the National Pork Board. The authors thank Nancy Upchurch, Pravina Kitikoon, Shan Yu, Margaret Aleff and Gavin Yager for technical assistance.

References Ameri-Mahabadi, M., Zhou, E.M., Hsu, W.H., 2005, Comparison of two swine Mycoplasma hyopneumoniae enzyme-linked immunosorbent assays for detection of antibodies from vaccinated pigs and field serum samples. J. Vet. Diagn. Invest. 17, 61-64. Asai, T., Okada, M., Ono, M., Irisawa, T., Mori, Y., Yokomizo, Y., Sato, S., 1993, Increased levels of tumor necrosis factor and interleukin 1 in bronchoalveolar lavage fluids from pigs infected with Mycoplasma hyopneumoniae. Vet. Immunol. Immunopathol. 38, 253-260. Bereiter, M., Young, T.F., Joo, H.S., Ross, R.F., 1990, Evaluation of the ELISA and comparison to the complement fixation test and radial immunodiffusion enzyme assay for detection of antibodies against Mycoplasma hyopneumoniae in swine serum. Vet. Microbiol. 25, 177-192. Erlandson, K.R., Evans, R.B., Thacker, B.J., Wegner, M.W., Thacker, E.L., 2005, Evaluation of three serum antibody enzyme-linked immunosorbent assays for Mycoplasma hyopneumoniae. J. Swine Health Prod. 13, 198-203. Friis, N.F., 1975, Some recommendations concerning primary isolation of Mycoplasma suipneumoniae and Mycoplasma flocculare a survey. Nord. Vet. Med. 27, 337-339. Holtkamp, D., Rotto, H., Parker, S. 2007. Costs of Swine Disease: Impacts of Feed Efficiency. In 15th Annual Swine Disease Conference for Swine Practitioners (Ames, IA), pp. 91-97. Lorenzo, H., Quesada, O., Assuncao, P., Castro, A., Rodriguez, F., 2006, Cytokine expression in porcine lungs experimentally infected with Mycoplasma hyopneumoniae. Vet. Immunol. Immunopathol. 109, 199-207. Madsen, M.L., Oneal, M.J., Gardner, S.W., Strait, E.L., Nettleton, D., Thacker, E.L., Minion, F.C., 2007, Array-based genomic comparative hybridization analysis of field strains of Mycoplasma hyopneumoniae. J. Bacteriol. 189, 7977-7982. Mengeling, W.L., Lager, K.M., Vorwald, A.C., 1995, Diagnosis of porcine reproductive and respiratory syndrome. J. Vet. Diagn. Invest. 7, 3-16.

Meyns, T., Maes, D., Calus, D., Ribbens, S., Dewulf, J., Chiers, K., de Kruif, A., Cox, E., Decostere, A., Haesebrouck, F., 2007, Interactions of highly and low virulent Mycoplasma hyopneumoniae isolates with the respiratory tract of pigs. Vet. Microbiol. 120, 87-95. Muneta, Y., Minagawa, Y., Shimoji, Y., Nagata, R., Markham, P.F., Browning, G.F., Mori, Y., 2006, IL-18 expression in pigs following infection with Mycoplasma hyopneumoniae. J. Interferon. Cytokine. Res. 26, 637-644. Ross, R.F., Zimmermann-Erickson, B.J., Young, T.F., 1984, Characteristics of protective activity of Mycoplasma hyopneumoniae vaccine. Am. J. Vet. Res. 45, 1899-1905. 34

Sambrook, J., David, W.R., 2001, Purification of nucleic acids, In: Argentine, J. (Ed.) Molecular Cloning: a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, pp. A8.9-A8.15. Stakenborg, T., Vicca, J., Butaye, P., Imberechts, H., Peeters, J., De Kruif, A., Haesebrouck, F., Maes, D., 2006, A multiplex PCR to identify porcine mycoplasmas present in broth cultures. Vet. Res. Commun. 30, 239-247. Stemke, G.W., Robertson, J.A., 1982, Comparison of two methods for enumeration of mycoplasmas. J. Clin. Microbiol. 16, 959-961. Thacker, E.L., Halbur, P.G., Ross, R.F., Thanawongnuwech, R., Thacker, B.J., 1999, Mycoplasma hyopneumoniae potentiation of porcine reproductive and respiratory syndrome virus-induced pneumonia. J. Clin. Microbiol. 37, 620-627. Thacker, E.L., Thacker, B.J., Kuhn, M., Hawkins, P.A., Waters, W.R., 2000a, Evaluation of local and systemic immune responses induced by intramuscular injection of a Mycoplasma hyopneumoniae bacterin to pigs. Am. J. Vet. Res. 61, 1384-1389. Thacker, E.L., Thacker, B.J., Young, T.F., Halbur, P.G., 2000b, Effect of vaccination on the potentiation of porcine reproductive and respiratory syndrome virus (PRRSV)- induced pneumonia by Mycoplasma hyopneumoniae. Vaccine 18, 1244-1252. Vicca, J., Stakenborg, T., Maes, D., Butaye, P., Peeters, J., de Kruif, A., Haesebrouck, F., 2003, Evaluation of virulence of Mycoplasma hyopneumoniae field isolates. Vet. Microbiol. 97, 177-190. Zielinski, G.C., Ross, R.F., 1990, Effect of growth in cell cultures and strain on virulence of Mycoplasma hyopneumoniae for swine. Am. J. Vet. Res. 51, 344-348.

CHAPTER 3. IMPACT OF GENETIC DIVERSITY AMONG MYCOPLASMA HYOPNEUMONIAE FIELD ISOLATES ON THE DEVELOPMENT OF TWO REAL-TIME ASSAYS

A paper submitted to the Journal of Clinical Microbiology

Erin L. Strait1, Melissa L. Madsen1, F. Chris Minion1, Jane Christopher-Hennings2, Matthew Dammen2, Katherine R. Jones1, and Eileen L. Thacker1

1Department of Veterinary Microbiology and Preventive Medicine, Iowa State University, Ames, IA 2Veterinary Science Department, South Dakota State University, Brookings, SD

Abstract Mycoplasma hyopneumoniae is an important cause of pneumonia in pigs around the world, but confirming its presence (or absence) in pigs can be difficult. Culture for diagnosis is impractical, and seroconversion is often delayed after natural infection, limiting the use of serology. Numerous PCR assays for the detection of M. hyopneumoniae have been developed targeting several different genes. Recently, genetic diversity among strains of M. hyopneumoniae has been demonstrated. The effect of this diversity on the accuracy and sensitivity of the M. hyopneumoniae PCR assays could result in false-negative results in current PCR tests. In this study a panel of isolates of M. hyopneumoniae, Mycoplasma flocculare, Mycoplasma hyorhinis and Mycoplasma hyosynoviae were tested with a number of M. hyopneumoniae-specific PCR assays. Some M. hyopneumoniae PCR assays tested did not detect all isolates of M. hyopneumoniae. To increase efficiency of PCR testing, two new real-time PCR assays were developed that are specific and detect all of the M. hyopneumoniae isolates used in this study.

Introduction Mycoplasma hyopneumoniae is the causative agent of enzootic pneumonia and an integral component of the porcine respiratory disease complex (PRDC). Considered one of the most important sources of disease-associated losses in swine production, M.

hyopneumoniae is also one of the most difficult to detect. This organism is difficult to isolate in pure culture because it is easily overgrown by other contaminating bacteria; therefore, culture is generally not attempted. In addition, seroconversion to M. hyopneumoniae is often slow within a herd and can vary considerably between pigs, making the use of ELISAs less effective. A number of PCR assays have been developed and reported to be both sensitive and specific for M. hyopneumoniae (4, 6, 11, 19, 20, 22, 24, 26). As a result, this technique is now among the most widely used for detection of M. hyopneumoniae in pigs. While the development of PCR assays has greatly enhanced our ability to detect M. hyopneumoniae, genetic variability of the organism (1, 5, 12, 14, 15, 21) can affect detection by PCR. It is not known what affect this heterogeneity has on the sensitivity of the different assays. Real-time PCR has many advantages over traditional PCR assays including less time to obtain quantitative results and a decrease in environmental contamination. Together, these traits make real-time PCR assays preferred in diagnostic and research settings. A real-time assay has not yet been described that is based on a conserved target common among isolates of M. hyopneumoniae. The two currently published M. hyopneumoniae-specific real-time assays had to be used in combination in order to detect all M. hyopneumoniae field samples in a Swiss collection of isolates (6). In addition, many of the previously published PCR assays were validated using only a small number of mycoplasma isolates. Therefore, the objectives of this study were to measure the ability of several M. hyopneumoniae-specific PCR assays to detect a collection of isolates of M. hyopneumoniae and also to identify new targets for real-time PCR assays for M. hyopneumoniae.

Materials and Methods Mycoplasmas. For this study four panels of commonly isolated mycoplasmas of swine were assembled including: 36 isolates of M. hyopneumoniae, 10 isolates of Mycoplasma flocculare, and 8 isolates each of Mycoplasma hyorhinis and Mycoplasma hyosynoviae. Also included were single isolates of several infrequently identified mycoplasmas and acholeplasmas of swine and Mycoplasma bovis from cattle. Clinical samples were obtained from pigs experimentally inoculated with different isolates of M. hyopneumoniae as previously described (28) and from uninfected control pigs. Samples were obtained 28 days- post-infection at necropsy (28). Cultures from a number of other swine bacterial pathogens were also evaluated (Table 3.1). Genomic DNA was obtained from each of the samples using the QIAamp DNA mini kit (Qiagen, Valencia, Calif.). PCR assays. A list of currently published PCR assays was identified (Table 3.2). These assays were then used to test DNA from each of the isolates in Table 3.1. Some assays were

performed as previously published (6, 11, 20) while others (4, 19, 22, 23, 26) were modified by substituting Master Amp E buffer (Ambion, Austin, Tex.) for the published buffer with their published cycling parameters in reactions consisting of 1x Master Amp E, 0.4 mM of each forward and reverse primer, 2.5U Taq (New England Biolabs, Inc., Ipswich, Maine), template containing 1-10 ng of DNA, and quantity sufficient of PCR-grade water to a final volume of 25 µl. The primer pairs from nested-set PCRs were run as individual assays (not as nested sets) to independently evaluate the ability of each primer pair to form products. M. hyopneumoniae mhp165 real-time PCR. The primers and probe for a real-time PCR assay targeting the mhp165 gene of M. hyopneumoniae (Genbank accession AE017332, M. hyopneumoniae 232 complete genome: base pairs 195124-201267) were designed using PrimerQuest software (Integrated DNA Technologies, Iowa City, Ia.) (Table 3.3: Mhp165 F, Mhp165 R, Mhp165 P). Three isolates of M. hyopneumoniae were sequenced in the region of the real-time product to evaluate sequence conservation in M. hyopneumoniae (Table 3.3: Mhp165 seqF and Mhp165 seqR). A panel of forward and reverse primers was then designed (Table 3.3: Mhp165 F g-a, Mhp165 R t-c, Mhp165 R t-c g-a, Mhp165 R g-a) based on the sequence results (Fig. 3.1). The primers were then tested for their ability to detect the isolates possessing single nucleotide polymorphisms (SNPs). One set of forward and reverse primers (Table 3.3: Mhp165 F and Mhp165 R) was selected and shown to be unique to M. hyopneumoniae by BLAST analysis. This set was further tested for specificity using a SYBR green assay performed as described by the manufacturer (QuantiTect SYBR® Green PCR Kit, Qiagen) with DNA from M. hyopneumoniae, M. flocculare, M. hyorhinis and M. hyosynoviae. All real-time assays were performed on a Rotor-Gene RG-3000 (Corbett Research, San Francisco, California). A checkerboard assay of primers at 100 nM, 300 nM and 900 nM versus probe concentrations at 50 nM, 150 nM and 250 nM was used to optimize the real-time protocol using the Universal PCR master mix (Applied Biosystems, Foster City, Calif.) according to the manufacturer’s directions. The following cycling parameters were used: 50ºC, 2 min; 95 ºC, 10 min; 40 cycles of 95 ºC, 15 sec, then 60ºC, 1 min. Template DNA from four different isolates of M. hyopneumoniae was utilized in replicates of the checkerboard assay. The detection limit of the finalized assay was tested using a dilution series (10 ng/µl to 1 fg/µl) of chromosomal DNA from M. hyopneumoniae strain 232. The assay was also compared to a nested PCR targeting the mhp165 gene that includes an inner set of primers (8) to a previously published outer set (19) (Table 3.3: Mhp165 outerF, Mhp165 outerR; Mhp165 innerF, Mhp165 innerR). Using the GeneAmp PCR core kit (Applied Biosystems) the following reaction was employed for both the inner and outer

Table 3.1. Bacteria used in this study. No. M. hyopneumoniae isolates Source Year of Origin Isolation/Receipt* 1 J ATCC 25934, type strain ATCC 2 232-2A3 (pig passage strain 11) Iowa State Univ. Iowa 3 37-9 Iowa State Univ. 1989 Kentucky 4 96MP0001 Iowa State Univ. 1996 5 96MP0002 Iowa State Univ. 1996 6 05MP0601 Iowa State Univ. 2005 Colorado 7 06MP0001D Iowa State Univ. 2006 8 3-14 Iowa State Univ. 1989 Michigan 9 00MP1301 Iowa State Univ. 2000 Wisconsin 10 05MP2301 Iowa State Univ. 2005 Minnesota 11 95MP1501 Iowa State Univ. 1995 Iowa 12 95MP1502 Iowa State Univ. 1995 Iowa 13 97MP0001 Iowa State Univ. 1997

14 95MP1503 Iowa State Univ. 1995 Iowa 38

15 95MP1511 Iowa State Univ. 1995 Iowa 16 95MP1504 Iowa State Univ. 1995 Iowa 17 95MP1505 Iowa State Univ. 1995 Iowa 18 95MP1506 Iowa State Univ. 1995 Iowa 19 95MP1507 Iowa State Univ. 1995 Iowa 20 95MP1509 Iowa State Univ. 1995 Iowa 21 95MP1510 Iowa State Univ. 1995 Iowa 22 00MP0001 Iowa State Univ. 2000 23 00MP0002 Iowa State Univ. 2000 24 00MP0003 Iowa State Univ. 2000 25 05MP2302A Iowa State Univ. 2005 Minnesota 26 05MP2303 Iowa State Univ. 2005 Minnesota 27 06MP0002 Iowa State Univ. 2006 28 06MP2501 Iowa State Univ. 2006 Missouri 29 00MP1502 Iowa State Univ. 2000 Iowa 30 P-1814-10 C. Armstrong, Pursue Univ. 1982* 31 P-5398-1 C. Armstrong, Purdue Univ. 1982* 32 P-5782 C. Armstrong, Purdue Univ. 1982*

Table 3.1 Continued. 33 P-6053-2 C. Armstrong, Purdue Univ. 1982* 34 P-11318-6 C. Armstrong, Purdue Univ. 1982* 35 P-12895-2 C. Armstrong, Purdue Univ. 1982* 36 P-13129-6 C. Armstrong, Purdue Univ. 1982* M. flocculare isolates 1 Ms42 27399, type strain ATCC 2 94MF1501 Iowa State Univ. 1994 Iowa 3 93MF0001 Iowa State Univ. 1993 4 P-6526 C. Armstrong, Purdue Univ. 1982* 5 P-9020-3 C. Armstrong, Purdue Univ. 1982* 6 P-10019 C. Armstrong, Purdue Univ. 1982* 7 P-10020 C. Armstrong, Purdue Univ. 1982* 8 P-18507-3 C. Armstrong, Purdue Univ. 1982* 9 Mp611 N. Friis, Denmark 1985*

10 Mp674 N. Friis, Denmark 1985* 39

11 Mp703 N. Friis, Denmark 1985* 12 02MF1501 Iowa State Univ. 2002 Iowa M. hyorhinis isolates 1 7 R.F. Ross, ISU 2 SK76C R.F. Ross, ISU 3 97MR3301 Iowa State Univ. 1997 North Carolina 4 97MR0001 Iowa State Univ. 1997 5 06MR0001 Iowa State Univ. 2006 6 06MR3501 Iowa State Univ. 2006 Ohio 7 97MR0002 Iowa State Univ. 1997 8 97MR1501 Iowa State Univ. 1997 Iowa 9 97MR0003 Iowa State Univ. 1997 10 04MR0001 Iowa State Univ. 2004 M. hyosynoviae isolates 1 S16 R.F. Ross, Iowa State Univ. 2 91MS0001 Iowa State Univ. 1991 3 94MS0001 Iowa State Univ. 1994 4 94MS0002 Iowa State Univ. 1994 5 95MS0001 Iowa State Univ. 1995

Table 3.1 Continued. 6 95MS0002 Iowa State Univ. 1995 7 97MS0002 Iowa State Univ. 1997 8 97MS0003 Iowa State Univ. 1997 9 91MS0002 Iowa State Univ. 1991 10 06MS1501 Iowa State Univ. 2006 Iowa 11 06MS2701 Iowa State Univ. 2006 Nebraska 12 97MS0001 Iowa State Univ. 1997 Additional bacteria 1 Mycoplasma arginini R.F. Ross, Iowa State Univ. 2 Mycoplasma hyopharyngis R.F. Ross, Iowa State Univ. 3 Mycoplasma salivarium R.F. Ross, Iowa State Univ. 4 Mycoplasma buccale R.F. Ross, Iowa State Univ. 5 Mycoplasma bovis R.F. Ross, Iowa State Univ. 40 6 Acholeplasma granularum R.F. Ross, Iowa State Univ.

7 Acholeplasma laidlawii B R.F. Ross, Iowa State Univ. 8 Actinobacillus pleuropneumoniae R.F. Ross, Iowa State Univ. 9 Arcanobacter pyogenes R.F. Ross, Iowa State Univ. 10 Bordetella bronchiseptica R. Griffith, Iowa State Univ. 11 Haemophilus parasuis R.F. Ross, Iowa State Univ. 1998 12 Erysipelothrix rhusiopathiae R. Griffith, Iowa State Univ. 1989 13 Pasteurella multocida (group A) R. Griffith, Iowa State Univ. 1987 14 Salmonella cholerasuis (group C1) R. Griffith, Iowa State Univ. 2005 15 Staphylococcus (pig tonsil) R. Griffith, Iowa State Univ. 16 Streptococcus suis type 1 R. Griffith, Iowa State Univ. 1986 17 Streptococcus suis type 2 R.F. Ross, Iowa State Univ. 2005 *Refers to year isolate was received from outside source. For these isolates, the year of isolation is unknown.

Table 3.2. Summary of results for PCR assays tested against M. hyopneumoniae isolates. PCR assays M. hyopneumoniae isolates A B C D E F G H I J K L M N O J ATCC 25934, type strain + + + - - + + + + + + + + + + 232-2A3 (pig passage of strain 11) + + + + + + + + + + + + + + + 37-9 + + + + + + + + + + + + + + + 96MP0001 - - - - - + + + + + + + - + + 96MP0002 + + + - - + + + + + + + + + + 05MP0601 + + + + + + + + + + + + + + + 06MP0001D + + + + + + + + + + + + + + + 3-14 + + + + + + + + + + + + + + + 00MP1301 + + + + + + + + + + + + - + + 05MP2301 + + + - - + + + + + + + + + + 95MP1501 - - - + + + + + + + + + + + + 95MP1502 + + + + + + + + + + + + + + + 97MP0001 + + + + + + + + + + + + + + + 95MP1503 + + + - + + + + + + + + + + + 95MP1511 + + + + + + + + + + + + + + + 95MP1504 + + + + + + + + + + + + + + + 95MP1505 + + + + + + + + + + + + + + + 95MP1506 + + + + + + + + + + + + + + + 95MP1507 + + + + - + + + + + + + + + + 95MP1509 + + + + + + + + + + + + + + + 95MP1510 + + + + + + + + + + + + + + + 00MP0001 + + + + + + + + + + + + + + + 00MP0002 + + + + + + + + + + + + + + + 00MP0003 + + + + + + + + + + + + + + + 05MP2302A + + + - - + + + + + + + + + + 05MP2303 + + + + + + + + + + + + + + + 06MP0002 + + + + + + + + + + + + - + + 06MP2501 + + + - - + + + + + + + + + + 00MP1502 + + + + + + + + + + + + - + + P-1814-10 - - - + + + + + + + + + - + + P-5398-1 + + + - - + + + + + + + - + + P-5782 - - - + + + + + + + + + + + + P-6053-2 + + + - - + + + + + + + - + + P-11318-6 + + + + + + + + + + + + - + + P-12895-2 + + + + + + + + + + + + - + + P-13129-6 + + + + + + + + + + + + + + + A, REP real-time PCR (6); B, outer nested product (22); C, inner nested product (22); D, ABC real-time PCR (6); E, inner nested product (26); F, multiplex assay (20); G, outer nested product (19); H, inner nested product (8); I, outer nested product (4, 13); J, inner nested product (4); K, (23); L, outer nested product (2, 11); M, inner nested product (11); N, mhp165 real-time PCR (this study); O, mhp183 real- time PCR (this study).

Table 3.3. List of primer and probe sequences used in this study. Primer/probe Product name* Sequence size Mhp165 seqF 5'-CCGGGCAATTCCAAGAGTTATTCAGG-3' Mhp165 seqR 5'-TTTGTGCTTCTAAACTAGGCCCTGC-3' 529 bp Mhp165 F 5'-TGCCCAGGATATTTCCGATCCAGA-3' Mhp165 R 5'-AGACCTGAAGAACGTGCATGGAGA-3' Mhp165 F g-a 5'-TGCCCAAGATATTTCCGATCCAGA-3' Mhp165 R t-c 5'-AGACCCGAAGAACGTGCATGGAGA-3' Mhp165 R t-c g-a 5'-AGACCCGAAGAACGTGCATGAAGA-3' Mhp165 R g-a 5'-AGACCTGAAGAACGTGCATGAAGA-3' Mhp165 P 5'-Cy5-GCGATCTCAACAAATACCGGGATTGGT-BHQ_2-3' 132 bp Mhp165 outerF 5'-TAGAAATGACTGGCAGACAA-3' Mhp165 outerR 5'-GAGGCTTTGATTTTGGAGTC-3' 835 bp Mhp165 innerF 5'-GATGCAGCAGGAAACTTAC-3' Mhp165 innerR 5'-CTCTGTTATCGGAGTTAG-3' 628 bp Mhp183 F 5'-CCAGAACCAAATTCCTTCGCTG-3' Mhp183 R 5'-ACTGGCTGAACTTCATCTGGGCTA-3' Mhp183 P 5'-6_FAM-AGCAGATCTTAGTCAAAGTGCCCGTG-BHQ_1-3' 90 bp * F=forward, R=reverse, P=probe

AE017332 GAATATGGTC ATCACCAAAC ACTAAATTAT GCCCAGGATA TTTCCGATCC AGATTCAAGT AE017244 ********** ********** ********** ********** ********** ********** AE017243 ********** ********** ********** ********** ********** ********** 96MP0001 ********** ********** ********** *****A**** ********** ********** 95MP1509 ********** ********** ********** *****A**** ********** ********** 06MP2501 ********** ********** ********** *****A**** ********** **********

AE017332 ATTATTCTTG ATGCGATCTC AACAAATACC GGGATTGGTT TGCAATCTTT TTATAATTTA AE017244 ********** ********** ********** ********** ********** ********** AE017243 ********** ********** ********** ********** ********** ********** 96MP0001 ********** ********** ********** ********** ********** ********** 95MP1509 ********** ********** ********** ********** ********** ********** 06MP2501 ********** ********** ********** ********** ********** **********

AE017332 GATGTTCTAA GAAAATATCT CCATGCACGT TCTTCAGGTC TAGACCTACG AAAAGCAGGG AE017244 ********** ********** ********** ********** ********** ***G**G*** AE017243 ********** ********** ********** *****G**** *****T**A* ********** 96MP0001 ********** ********** T********* *****G**** *****T**A* ********** 95MP1509 ********** ********** ********** ********** *****T**** ********** 06MP2501 ********** ********** ********** *****G**** *****T**A* ********** Figure 3.1. Sequence alignment of the mhp165 gene real-time PCR target region for isolates of M. hyopneumoniae compared to the consensus sequence from M. hyopneumoniae strain 232 (AE017332). Forward primer, Probe, Reverse primer, an * indicates homologous sequence.

43 primer, 2.5U Taq, 2 µl template (PCR product from the outer reaction is template for the inner “nested” reaction), and quantity sufficient of PCR-grade water to a final volume of 50 µl. The same cycling parameters were applied to both the outer and inner reactions: 94ºC, 3 min; 35 assays: 1x PCR buffer II, 2.5 mM MgCl2, 100 µM each dNTP, 0.4 mM each forward and reverse cycles of 94ºC, 1 min, 55ºC, 1 min and 72ºC, 2 min. Products from these reactions were analyzed on a 1.5% agarose gel. M. hyopneumoniae mhp183 real-time PCR. To identify conserved regions for primer and probe design, gene sequences for P97 (mhp183) from 17 different GeneBank records were aligned using Vector NTI Advance™ 10 (Invitrogen, Carlsbad, CA) (Fig. 3.2), and primers and probes were designed with Primer Express Ver. 2.0.0 (Applied Biosystems) (Table 3.3: Mhp183 F, Mhp183 R, and Mhp183 P). These primers and probes were shown to be unique to M. hyopneumoniae by a BLAST search. They were further tested for specificity with a SYBR green assay performed as described by the manufacturer (QuantiTect SYBR® Green PCR Kit, Qiagen) with DNA from M. hyopneumoniae, M. flocculare, M. hyorhinis and M. hyosynoviae. A checkerboard assay of primers versus probe concentrations was performed as described above. Template DNA from four different isolates of M. hyopneumoniae was utilized in replicates of the checkerboard assay. This assay was then tested against each of the bacterial DNA samples listed in Table 3.1. The detection level of the finalized assay was tested with a dilution series (10 ng/µl to 1 fg/µl) of chromosomal DNA of M. hyopneumoniae. mhp165 and mhp183 multiplex real-time PCR assay. A dilution series of chromosomal DNA of M. hyopneumoniae was tested in the mhp165 and mhp183 real-time PCR assays both individually and multiplexed, and with or without an exogenous internal positive control (EIPC) (Applied Biosystems) (data not shown). The EIPC was used to control for inhibition in the PCR reaction caused by the template sample. Reagents for the EIPC were diluted 2-fold from the manufacturer’s suggestion. DNA sequencing and analysis. The region of the M. hyopneumoniae genome containing the Artiushin et al. PCR target (2) and used by Kurth et al. in the outer pair of a nested PCR reaction (11), was sequenced in strains 232, 95MP1509, 95MP1505, 00MP1502, and 00MP1301. The location of the PCR target was identified by BLAST analysis of the M. hyopneumoniae genome sequence (16) with the sequence in Kurth et al. and was found to include the 5’ end of mhp023 and most of mhp024. Nucleotide sequence accession numbers. The nucleotide sequences in the mhp023-mhp024 chromosomal regions of strains 95MP1505, 95MP1509, 00MP1301 and 00MP1502 have been submitted to the GenBank database under accession numbers XX, XX, XX, and XX, respectively. The nucleotide sequences in the mhp165 chromosomal region of strains

95MP1509, 96MP0001 and 06MP2501 have been submitted to the GenBank database under accession numbers XX, XX, and XX, respectively.

AE017332 GAATTTAGCT TTGCAAAAGA TAGTTCAACT AATCAATATG TAAGTATCCA GAACCAAATT AE017244 ********** ********** ********** ********** ********** ********** AE017243 ********** ********** ********** ********** ********** ********** AY957500 ********** ********** ********** ********** ********** ********** AY512905 ********** ********** ********** ********** ********** ********** AY512904 ********** ********** ********** ********** ********** ********** AY512903 A********* ********** ********** ********** ********** ********** AY512902 ********** ********** ********** ********** ********** ********** AY512901 ********** ********** ********** ********** ********** ********** AY512900 ********** ********** ********** ********** ********** ********** AY512899 ********** ********** ********** ********** ********** ********** AY512898 A********* ********** ********** ********** ********** ********** AY512897 ********** ********** ********** ********** ********** ********** AY512896 ********** ********** ********** ********** ********** ********** AY512895 ********** ********** ********** ********** ********** ********** AY512894 ********** ********** ********** ********** ********** ********** U27294 A********* ********** ********** ********** ********** **********

AE017332 CCTTCGCTGT TTTTAAAAGC AGATCTTAGT CAAAGTGCCC GTGAAATTTT AGCTAGCCCA AE017244 ********** ********** ********** ********** ********** ********** AE017243 ********** ********** ********** ********** ********** ********** AY957500 *********G ********** ********** ********** ********** ********** AY512905 ********** ********** ********** ********** ********** ********** AY512904 ********** ********** ********** ********** ********** ********** AY512903 ********** ********** ********** ********** ********** ********** AY512902 ********** ********** ********** ********** ********** ********** AY512901 ********** ********** ********** ********** ********** ********** AY512900 ********** ********** ********** ********** ********** ********** AY512899 ********** ********** ********** ********** ********** ********** AY512898 ********** ********** ********** ********** ********** ********** AY512897 ********** ********** ********** ********** ********** ********** AY512896 ********** ********** ********** ********** ********** ********** AY512895 ********** ********** ********** ********** ********** ********** AY512894 ********** ********** ********** ********** ********** ********** U27294 ********** ********** ********** ********** ********** **********

AE017332 GATGAAGTTC AGCCAGTTAT TAACATTTTA AGATTAATGA AAAAAGATAA TTCTTCTTAT AE017244 ********** ********** ********** ********** ********** ********** AE017243 ********** ********** ********** ********** ********** ********** AY957500 ********** ********** ********** ********** ********** ********** AY512905 ********** ********** ********** ********** ********** ********** AY512904 ********** ********** ********** ********** ********** ********** AY512903 ********** ********** ********** ********** ********** ********** AY512902 ********** ********** ********** ********** ********** ********** AY512901 ********** ********** ********** ********** ********** ********** AY512900 ********** ********** ********** ********** ********** ********** AY512899 ********** ********** ********** ********** ********** ********** AY512898 ********** ********** ********** ********** ********** ********** AY512897 ********** ********** ********** ********** ********** ********** AY512896 ********** ********** ********** ********** ********** ********** AY512895 ********** ********** ********** ********** ********** ********** AY512894 ********** ********** ********** ********** ********** ********** U27294 ********** ********** ********** ********** ********** ********** Figure 3.2. Sequence alignment of the mhp183 gene real-time PCR target region for isolates of M. hyopneumoniae compared to the consensus sequence from M. hyopneumoniae strain 232 (AE017332). Forward primer, Probe, Reverse primer, an * indicates homologous sequence.

Results Mycoplasma PCR assay panel. Assays targeting the 16S ribosomal RNA gene (4, 20, 24) and the mhp165 gene (8, 19) positively identified all of the M. hyopneumoniae isolates used in this study. However, five of the M. hyopneumoniae-specific PCR assays did not detect every M. hyopneumoniae isolate in our collection (Table 3.2). These five assays targeted the following three genes: a putative ABC transporter (Genbank accession no. U02537) (6, 26), a repeated element (Genbank accession no. AF004388) (6, 22), and a target that spanned the hypothetical genes mhp023 and mhp024 (Genbank accession no. AE017332, base pairs 27057-28020) (11). With the Kurth et al. PCR assay, some strains of M. hyopneumoniae gave products with the outer primers, but failed on the inner primer pair. Real-time PCR assays. The SYBR green assay using primers for the mhp165 assay were specific only to M. hyopneumoniae, while the primers for the mhp183 assay produced a product in both M. hyopneumoniae and M. flocculare. When analyzed on a 3% agarose gel, the product in M. flocculare was smaller than the amplicon made in M. hyopneumoniae (data not shown), indicating that the targets are not identical in each species. The most sensitive combinations of primers and probes among the M. hyopneumoniae isolates tested for the mhp165 assay were: 900 nM forward primer, 900 nM reverse primer and 50 nM probe. Additionally, Mhp165 probes labeled with Cy3 or Cy5 were found to perform identically regardless of the label used (data not shown). For the mhp183 assay, the most sensitive combination was: 50 nM forward primer, 900 nM reverse primer, 50 nM probe. Both of the probe-based real-time PCR assays were specific to M. hyopneumoniae, detected all isolates of M. hyopneumoniae in our DNA collection and detected none of those in the M. flocculare, M. hyorhinis or M. hyosynoviae panels. In addition, none of the other bacteria tested were positive by either of these assays (100% specificity). Clinical samples taken 28 days post inoculation from M. hyopneumoniae infected pigs were also tested including nasal swabs, bronchial swabs and bronchial alveolar lavage fluid. All samples from all infected pigs were positive by both assays except for one nasal swab tested by the mhp183 assay (Table 3.4). DNA sequence results. DNA sequence analysis of the Kurth et al. PCR target was performed due to the failure of the inner product of the nested PCR to detect some strains of M. hyopneumoniae. Sequences from strains 232, 95MP1509, and 95MP1505 are nearly identical (data not shown), but sequences from strains 00MP1502 and 00MP1301 show three regions with only 80-87% homology to strain 232 and three regions with no significant homology including a break in the alignment of 00MP1502 and 00MP1301 to 232 at the site where the forward inner primer binds. Sequence of the inner-primer binding regions and a graphical representation of the Blast scores are shown in Figure 3.3.

Sequence analysis of the 529 base pair region containing the primer binding sites in mhp165 for five isolates of M. hyopneumoniae identified SNPs in the reverse and/or forward primers, while the site of probe binding was found to be homologous (Fig. 3.1). Real-time PCR analysis using DNA templates from isolates 96MP0001, 95MP1509, 06MP2501, and 232 was then performed using primers including each of the identified SNPs (primers for the 132 bp fragment, Table 3.3). Every primer pair was able to detect each isolate tested, but variation in the sensitivity among M. hyopneumoniae isolates was observed as a change in the Ct value among the different primer pairs (data not shown).

Table 3.4. Analysis of clinical samples by real-time PCR.

mhp165 real-time PCR results mhp183 real-time PCR results No. positive/No. of samples tested No. positive/No. of samples tested Challenge isolate NS BS BALF NS BS BALF 95MP1509 7/7 7/7 95MP1505 7/7 7/7 00MP1301 2/2 2/2 9/9 1/2 2/2 7/7 00MP1502 2/2 2/2 9/9 2/2 2/2 7/7 95MP1501 2/2 2/2 2/2 2/2 2/2 2/2 97MP0001 2/2 2/2 2/2 2/2 2/2 2/2 232 36/36 36/36 Unchallenged controls 0/33 0/33 NS = nasal swab; BS = bronchial swab; BALF = bronchial alveolar lavage fluid

Detection level. A 10-fold dilution series from 10 ng/µl to 1 fg/µl including two additional dilutions of 5 fg/µl and 2.5 fg/µl was made with DNA extracted from M. hyopneumoniae strain 232. One fg of chromosomal DNA is considered to be approximately one genome equivalent (11). The nested-set PCR based on the mhp165 gene detected DNA from the 10 ng/µl to 5 fg/µl dilutions. The mhp165 and mhp183 real-time PCR assays each detected DNA from the 10 ng/µl to 2.5 fg/µl dilutions (data not shown).

Figure 3.3 DNA sequence analysis of the Kurth et al. inner PCR product of strains 00MP1502 and 00MP1301. A. DNA sequence alignment of strains 232, 00MP1502, and 00MP1301in the regions of the inner PCR product primers Mhp3 and Mhp4. Identities are shown by the vertical lines. B. Graphical representation of the BLAST alignment of the 00MP1502 and 00MP1301 inner PCR product strain sequences with strain 232. Gray areas indicate poor or no alignment. White regions indicate homology with the percent identity indicated below. The region of the inner PCR product is indicated by the double arrow. The outer primers bind outside the indicated region, and DNA sequence was not obtained for that part of the chromosome.

Discussion Five PCR assays targeting three different genes of M. hyopneumoniae failed to detect all of the M. hyopneumoniae isolates in our collection. The two published real-time assays targeting the ABC and REP genes did not detect every isolate as had been previously reported (6). Additionally, the standard PCR assays on which these two real-time assays were based failed to detect the same isolates of M. hyopneumoniae in nearly every case (22, 26). Contrary to Dubosson et al. (6) who used the REP and ABC real-time assays in combination to detect all isolates in their study, there was one isolate (96MP0001, Table 3.1) in the current study that was not detected by either of these PCR assays. The Kurth et al. PCR assay, a nested assay targeting a unique hypothetical gene of M. hyopneumoniae (11), also failed to detect all isolates. The basis for this failure was genetic variability among the isolates resulting in deletions and rearrangements in the mhp024 gene. One of the PCR primer-binding sites was missing causing failure in the inner-pair reaction. This demonstrates the importance of evaluating each primer-pair individually across multiple isolates in nested-set PCR assays. Based on these PCR results we developed two new real-time PCR assays. The nested PCR assay to the gene mhp165 performed well against our panel of isolates, and therefore this gene was selected as a target for one of the new real-time assays. The function of the mhp165 gene

48 has not been identified, but it is a large gene (6140 base pairs) containing zinc finger, metalopeptidase, coiled-coiled, transmembrane, secretory signal, and ABC transporter sequence motifs. This information is available in the M. hyopneumoniae database (http://mycoplasma.genome.uab.edu/). A second real-time PCR assay developed in this study targets the gene for the P97 cilium adhesin (mhp183). The product of this well-characterized gene is important for adherence of M. hyopneumoniae to ciliated epithelium within the respiratory tract (9, 10, 27). Because it is thought to be necessary for virulence, this gene is likely to be present in all pathogenic isolates of M. hyopneumoniae. Additionally, this gene has been sequenced from a large number of isolates allowing identification of conserved regions. While a small target was produced with the M. flocculare template using the forward and reverse primer in a SYBR green assay, this product did not hybridize to our M. hyopneumoniae-specific probe. When M. flocculare DNA was added to a dilution series of M. hyopneumoniae DNA, no affect on sensitivity to the real-time assay was found (data not shown). Due to the restricted parameters used in primer and probe design programs for real-time PCR assays, there is often little choice in the area of the gene that can be targeted. This is particularly true for M. hyopneumoniae due to its low G+C content (28.6%)(16). Therefore, even though SNPs were identified in the primer binding sites for mhp165, this area was unable to be avoided because of probe design issues; however, the internal probe was conserved between these isolates. In addition to the SNPs identified in the previously sequenced strains: 232, J and 7228, three isolates from the M. hyopneumoniae isolate collection were sequenced and found to contain various SNPs in the primer-binding regions. More extensive sequencing of isolates was not performed due to the fact that all isolates in our panel were detected by this assay. The observed SNPs did have an affect on the sensitivity of the mhp165 assay. This could potentially affect the diagnostic ability of the mhp165 assay in the field. Additionally, this assay would not be as useful for quantitative comparisons of samples from different isolates as the mhp183 assay which appears to be conserved at the primer and probe binding sites. Both the mhp165 and mhp183 assays detected M. hyopneumoniae DNA in clinical samples collected from groups of pigs experimentally inoculated with M. hyopneumoniae isolates. One nasal swab sample was found to be negative by the mhp183 assay from a pig challenged with isolate 00MP1301. Performing PCR on samples from nasal swabs has previously been shown to have a lower detection rate than samples from bronchial swabs or bronchial alveolar lavage fluid (11). The nasal swab sample from the other pig challenged with isolate 00MP1301 and the bronchial swabs and bronchial alveolar lavage fluid samples from both pigs were all PCR- positive demonstrating the ability of the mhp183 assay to detect this isolate. The exogenous internal positive control included in the mhp183 assay gave a positive result in the nasal swab

49 sample that was negative (data not shown) indicating no sample inhibition occurred and that the negative result was more likely due to the low level of M. hyopneumoniae DNA present in the sample. The assays targeting the 16S ribosomal RNA gene detected all M. hyopneumoniae isolates (4, 20, 24), but many researchers are reluctant to target this gene due to its high level of conservation among bacterial species that may lead to false positive results. A number of mycoplasmas and related acholeplasmas that have received little research such as M. arginini, M. salivarium, M. hyopharyngis, A. laidlawii, and A. granularum and others have been isolated from pigs (3, 7, 17, 18). The prevalence of these organisms is unclear as is their potential for eliciting false-positive results in 16S-based PCR assays (or in PCR assays targeting other genes). Although three strains of M. hyopneumoniae have been sequenced (16, 25), there is little known concerning the extent of genetic variability between M. hyopneumoniae isolates, making it difficult to identify conserved targets for PCR assays. One recent study has shown a considerable amount of genetic diversity among field isolates (12). To improve the diagnostic ability to detect field isolates of M. hyopneumoniae, the two real-time PCR assays developed for mhp165 and mhp183 were multiplexed. An EIPC was included to control for sample inhibition from template contaminants leading to a false negative result. No decrease in sensitivity was found when comparing the individual assays to the multiplexed assay. Each of the two real-time assays was more sensitive than a nested-set PCR assay to the mhp165 gene (8). Noteworthy is the fact that the mhp165 and mhp183 assays described here detected all of the isolates tested. This is a significant improvement over previously described real-time PCR assays. Results from this study also support previously reported genetic diversity among isolates of M. hyopneumoniae. The extent of this variability and its affect on currently used diagnostic assays to accurately detect a wide range of M. hyopneumoniae field isolates has yet to be determined. The majority of M. hyopneumoniae isolates used in the present study came from pigs in the United States. A future study comparing these PCR assays against a more diverse international collection of M. hyopneumoniae isolates may be beneficial to further test their sensitivity and specificity.

Acknowledgements

This study was funded by a grant from the National Pork Board. The authors would like to thank Barbara Zimmerman Erickson and Linda Zeller for their assistance in acquiring cultures for this study and Nancy Upchurch for media preparation. The authors also thank Michael Carruthers for his technical assistance.

References 1. Artiushin, S., and F. C. Minion. 1996. Arbitrarily primed PCR analysis of Mycoplasma hyopneumoniae field isolates demonstrates genetic heterogeneity. Int. J. Syst. Bacteriol. 46:324-328. 2. Artiushin, S., L. Stipkovits, and F. C. Minion. 1993. Development of polymerase chain reaction primers to detect Mycoplasma hyopneumoniae. Mol. Cell. Probes 7:381-385. 3. Blank, W. A., B. Z. Erickson, and G. W. Stemke. 1996. Phylogenetic relationships of the porcine mycoplasmas Mycoplasma hyosynoviae and Mycoplasma hyopharyngis. Int. J. Syst. Bacteriol. 46:1181-1182. 4. Calsamiglia, M., C. Pijoan, and A. Trigo. 1999. Application of a nested polymerase chain reaction assay to detect Mycoplasma hyopneumoniae from nasal swabs. J. Vet. Diagn. Invest. 11:246-251. 5. de Castro, L. A., T. R. Pedroso, S. S. Kuchiishi, M. Ramenzoni, J. D. Kich, A. Zaha, M. H. Vainstein, and H. B. Ferreira. 2006. Variable number of tandem aminoacid repeats in adhesion-related CDS products in Mycoplasma hyopneumoniae strains. Vet. Microbiol. 116:258-269. 6. Dubosson, C. R., C. Conzelmann, R. Miserez, P. Boerlin, J. Frey, W. Zimmermann, H. Hani, and P. Kuhnert. 2004. Development of two real-time PCR assays for the detection of Mycoplasma hyopneumoniae in clinical samples. Vet. Microbiol. 102:55-65. 7. Erickson, B. Z., R. F. Ross, and J. M. Bove. 1988. Isolation of Mycoplasma salivarium from swine. Vet. Microbiol. 16:385-390. 8. Fraser, T., J. Christopher-Hennings, K. Rossow, E. Thacker, P. Ahrens, R. Chapin, A. Wasilk, and E. Nelson. 1999. Detection of Mycoplasma hyopneumoniae by nested PCR: A comparison of specimen type and extraction methods. p. 36. Abstr. 42nd Annual Meet. Am. Assoc. Vet. Lab. Diag., San Diego, CA. 9. Hsu, T., S. Artiushin, and F. C. Minion. 1997. Cloning and functional analysis of the P97 swine cilium adhesin gene of Mycoplasma hyopneumoniae. J. Bacteriol. 179:1317-1323. 10. Hsu, T., and F. C. Minion. 1998. Identification of the cilium binding epitope of the Mycoplasma hyopneumoniae P97 adhesin. Infect. Immun. 66:4762-4766. 11. Kurth, K. T., T. Hsu, E. R. Snook, E. L. Thacker, B. J. Thacker, and F. C. Minion. 2002. Use of a Mycoplasma hyopneumoniae nested polymerase chain reaction test to determine the optimal sampling sites in swine. J. Vet. Diagn. Invest. 14:463-469. 12. Madsen, M. L., M. J. Oneal, S. W. Gardner, E. L. Strait, D. Nettleton, E. L. Thacker, and F. C. Minion. 2007. Array-based genomic comparative hybridization analysis of field strains of Mycoplasma hyopneumoniae. J. Bacteriol. 189:7977-7982.

13. Mattsson, J. G., K. Bergstrom, P. Wallgren, and K. E. Johansson. 1995. Detection of Mycoplasma hyopneumoniae in nose swabs from pigs by in vitro amplification of the 16S rRNA gene. J. Clin. Microbiol. 33:893-897. 14. Mayor, D., J. Jores, B. M. Korczak, and P. Kuhnert. 2007. Multilocus sequence typing (MLST) of Mycoplasma hyopneumoniae: A diverse pathogen with limited clonality. Vet. Microbiol. (In press). 15. Mayor, D., F. Zeeh, J. Frey, and P. Kuhnert. 2007. Diversity of Mycoplasma hyopneumoniae in pig farms revealed by direct molecular typing of clinical material. Vet. Res. 38:391-398. 16. Minion, F. C., E. J. Lefkowitz, M. L. Madsen, B. J. Cleary, S. M. Swartzell, and G. G. Mahairas. 2004. The genome sequence of Mycoplasma hyopneumoniae strain 232, the agent of swine mycoplasmosis. J. Bacteriol. 186:7123-7133. 17. Orning, A. P., R. F. Ross, and M. F. Barile. 1978. Isolation of Mycoplasma arginini from swine and from a swine waste disposal system. Am. J. Vet. Res. 39:1169-1174. 18. Roberts, D. H., and T. W. Little. 1976. Isolation of Acholeplasma granularum from porcine faeces. Vet. Rec. 99:13. 19. Sorensen, V., P. Ahrens, K. Barfod, A. A. Feenstra, N. C. Feld, N. F. Friis, V. Bille- Hansen, N. E. Jensen, and M. W. Pedersen. 1997. Mycoplasma hyopneumoniae infection in pigs: duration of the disease and evaluation of four diagnostic assays. Vet. Microbiol. 54:23-34. 20. Stakenborg, T., J. Vicca, P. Butaye, H. Imberechts, J. Peeters, A. De Kruif, F. Haesebrouck, and D. Maes. 2006. A multiplex PCR to identify porcine mycoplasmas present in broth cultures. Vet. Res. Commun. 30:239-247. 21. Stakenborg, T., J. Vicca, D. Maes, J. Peeters, A. de Kruif, F. Haesebrouck, and P. Butaye. 2006. Comparison of molecular techniques for the typing of Mycoplasma hyopneumoniae isolates. J. Microbiol. Methods 66:263-275. 22. Stark, K. D., J. Nicolet, and J. Frey. 1998. Detection of Mycoplasma hyopneumoniae by air sampling with a nested PCR assay. Appl. Environ. Microbiol. 64:543-548. 23. Stemke, G. W. 1997. Gene amplification (PCR) to detect and differentiate mycoplasmas in porcine mycoplasmal pneumonia. Lett. Appl. Microbiol. 25:327-330. 24. Stemke, G. W., R. Phan, T. F. Young, and R. F. Ross. 1994. Differentiation of Mycoplasma hyopneumoniae, M flocculare, and M hyorhinis on the basis of amplification of a 16S rRNA gene sequence. Am. J. Vet. Res. 55:81-84. 25. Vasconcelos, A. T., H. B. Ferreira, C. V. Bizarro, S. L. Bonatto, M. O. Carvalho, P. M. Pinto, D. F. Almeida, L. G. Almeida, R. Almeida, L. Alves-Filho, E. N. Assuncao, V. A. Azevedo, M. R. Bogo, M. M. Brigido, M. Brocchi, H. A. Burity, A. A. Camargo, S. S.

Camargo, M. S. Carepo, D. M. Carraro, J. C. de Mattos Cascardo, L. A. Castro, G. Cavalcanti, G. Chemale, R. G. Collevatti, C. W. Cunha, B. Dallagiovanna, B. P. Dambros, O. A. Dellagostin, C. Falcao, F. Fantinatti-Garboggini, M. S. Felipe, L. Fiorentin, G. R. Franco, N. S. Freitas, D. Frias, T. B. Grangeiro, E. C. Grisard, C. T. Guimaraes, M. Hungria, S. N. Jardim, M. A. Krieger, J. P. Laurino, L. F. Lima, M. I. Lopes, E. L. Loreto, H. M. Madeira, G. P. Manfio, A. Q. Maranhao, C. T. Martinkovics, S. R. Medeiros, M. A. Moreira, M. Neiva, C. E. Ramalho-Neto, M. F. Nicolas, S. C. Oliveira, R. F. Paixao, F. O. Pedrosa, S. D. Pena, M. Pereira, L. Pereira-Ferrari, I. Piffer, L. S. Pinto, D. P. Potrich, A. C. Salim, F. R. Santos, R. Schmitt, M. P. Schneider, A. Schrank, I. S. Schrank, A. F. Schuck, H. N. Seuanez, D. W. Silva, R. Silva, S. C. Silva, C. M. Soares, K. R. Souza, R. C. Souza, C. C. Staats, M. B. Steffens, S. M. Teixeira, T. P. Urmenyi, M. H. Vainstein, L. W. Zuccherato, A. J. Simpson, and A. Zaha. 2005. Swine and poultry pathogens: the complete genome sequences of two strains of Mycoplasma hyopneumoniae and a strain of Mycoplasma synoviae. J. Bacteriol. 187:5568-5577. 26. Verdin, E., C. Saillard, A. Labbe, J. M. Bove, and M. Kobisch. 2000. A nested PCR assay for the detection of Mycoplasma hyopneumoniae in tracheobronchiolar washings from pigs. Vet. Microbiol. 76:31-40. 27. Zhang, Q., T. F. Young, and R. F. Ross. 1995. Identification and characterization of a Mycoplasma hyopneumoniae adhesin. Infect. Immun. 63:1013-1019. 28. Zielinski, G. C., and R. F. Ross. 1990. Effect of growth in cell cultures and strain on virulence of Mycoplasma hyopneumoniae for swine. Am. J. Vet. Res. 51:344-348.

CHAPTER 4. COMPARISON OF MYCOPLASMA HYOPNEUMONIAE- SPECIFIC ELISAS INCLUDING THE EFFECT OF FIELD ISOLATE VARIABILITY

A paper to be submitted to Clinical and Vaccine Immunology

E. L. Strait, B. Z. Erickson, E. L. Thacker

Dept. of Veterinary Microbiology and Preventive Medicine, College of Veterinary Medicine, Iowa State University, Ames, IA.

Abstract The objective of this study was to assess the sensitivity and specificity of three Mycoplasma hyopneumoniae-specific ELISAs and examine the effect of field isolate variability on detection by the Tween 20 ELISA, the HerdChek® M. hyopneumoniae ELISA (IDEXX Laboratories, Westbrook, ME), and the IDEIA™ M. hyopneumoniae EIA KIT (Oxoid, Ely, Cambridgeshire, UK, formerly DAKO™ M. hyopneumoniae ELISA). The effect of field isolate variability on detection by these assays was also examined. A time-course study was performed in groups of pigs challenged with M. hyopneumoniae, Mycoplasma flocculare, Mycoplasma hyorhinis or Mycoplasma hyosynoviae. Four groups received one of four genetically distinct field isolates of M. hyopneumoniae. Two groups received different isolates of M. flocculare, and one group each received M. hyorhinis or M. hyosynoviae. Two pigs in each group were necropsied at 28, 70 and 95 days-post-challenge (DPC). All sera were tested by the three ELISAs. All three ELISAs were specific for M. hyopneumoniae. The Oxoid ELISA was the most sensitive assay, detecting pigs earlier (14 DPC) and more consistently than either the Tween 20 or IDEXX ELISAs, which did not detect any positives until 28 DPC. The Tween 20 and IDEXX ELISAs overall performed similarly to each other, but differed 24% of the time on individual results from 28-95 DPC. In addition, the Tween 20 and IDEXX ELISAs detected lower levels of antibodies in pigs inoculated with one of the M. hyopneumoniae isolates suggesting that differences in antibody production to field isolates affects sensitivity in these two ELISAs.

Introduction Mycoplasma hyopneumoniae, the causative agent of enzootic pneumonia of swine, is a source of significant economic loss in the swine industry today. It is also a key component of the porcine respiratory disease complex. While M. hyopneumoniae is currently ubiquitous in the countries where it is found, many producers are working to eradicate this organism from their production systems. Therefore there is a need to evaluate the presence of M. hyopneumoniae within swine herds. Serology is the most commonly used tool to monitor herd status due to its low cost and the relative ease of sample collection for these assays. Current M. hyopneumoniae ELISAs include the 232 Tween 20 ELISA (made using antigen from M. hyopneumoniae strain 232) (4, 14) and the HerdChek® M. hyopneumoniae Antibody Test Kit (IDEXX Laboratories, Westbrook, ME), which are both indirect ELISAs, and a competitive-inhibition ELISA: the IDEIA™ M. hyopneumoniae EIA KIT (Oxoid, Ely, Cambridgeshire, UK, formerly the DAKO™ M. hyopneumoniae ELISA). The antibodies detected in the indirect assays recognize numerous antigens primarily associated with the cell membrane, while the Oxoid ELISA is designed to detect antibodies to a single intracellular protein of M. hyopneumoniae. A number of studies have evaluated the sensitivity and specificity of these three ELISAs (1, 7, 8, 18-20). Some of the sera used for these comparisons were field samples, while others came from challenge studies. The true infection status of sera collected from pigs in the field can not always be demonstrated, making it difficult to definitively measure specificity using these samples. Experimentally-derived sera used a previous comparison of these ELISAs to each other came from pigs challenged with M. hyopneumoniae strain 232 (7), the isolate that the 232 Tween 20 is based upon. While sera from experimentally-challenged pigs can be used to compare the sensitivity of the different ELISAs to serum antibodies, they do not effectively evaluate the ability of the ELISAs to detect antibodies against a variety of isolates of M. hyopneumoniae and do not measure specificity. Previous studies comparing different isolates of M. hyopneumoniae using SDS-PAGE and western blots have demonstrated diversity in protein expression (3, 17). Some of this variation may be due to alterations in post-translational processing of surface antigens (5, 6) In addition, western blot analyses demonstrated that pigs develop varying antibody profiles following infection with M. hyopneumoniae in the field (2). The effect of M. hyopneumoniae protein variation on the ability of ELISAs to detect antibodies in a diagnostic setting is not fully known. Pigs are the host for a number of mycoplasmas including Mycoplasma flocculare, Mycoplasma hyorhinis and Mycoplasma hyosynoviae. Some concern exists for the potential of these mycoplasmas to cause false-positives in M. hyopneumoniae-specific ELISAs, 55

especially M. flocculare which is the most antigenically similar to M. hyopneumoniae (4). As a result there is a need to evaluate M. hyopneumoniae-specific ELISAs for their sensitivity and specificity using serology from pigs with known infection status in relation to other species of mycoplasmas in addition to genetically different field isolates of M. hyopneumoniae. The objectives of this study were to evaluate the ability of three serological assays to detect antibodies to isolates of M. hyopneumoniae at various time-points following infection, and assess their sensitivity and specificity to other mycoplasma species commonly found in pigs. The virulence of two of the M. hyopneumoniae isolates used in the current study: 00MP1301 and 00MP1502, have been established in a previous study where 00MP1301 was found to be significantly more virulent than isolate 00MP1502 (21). All four of the M. hyopneumoniae isolates in the current study have been compared to strain 232 by microarray analysis and found to be genetically unique from 232 and from each other (12).

Materials and Methods

Animals, housing and experimental design. Forty-eight, 10-12 day-old pigs obtained from a commercial herd negative for M. hyopneumoniae were divided into eight groups of six pigs housed in separate rooms in a biosecure animal isolation facility at Iowa State University. At approximately one month of age, the pigs were challenged on three consecutive days with log-phase culture (10 ml intratracheally for M. hyopneumoniae (16) and M. flocculare groups, or 5 ml intranasally for M. hyorhinis and M. hyosynoviae groups). Mycoplasma hyopneumoniae and M. flocculare were grown in Friis media (9) in a shaking waterbath at 37°C, 50 rpm; M. hyorhinis was grown in BHI media (15), stationary at 37°C; and M. hyosynoviae was grown in D-TS media (15), stationary at 37°C. There were four M. hyopneumoniae groups (isolates: 95MP1501, 97MP0001, 00MP1301 and 00MP1502), two M. flocculare groups (isolates: 94MF1501 and 02MF1501) and one group each of M. hyorhinis (isolate: SK76C) and M. hyosynoviae (isolate: 97MS0003). Two pigs were necropsied at each of the following time points: 28, 70 and 96 days-post-challenge (DPC). One pig in each of the 00MP1301 and 02MF1501groups died from causes unrelated to the study; therefore, in these groups one pig rather than two were necropsied at 70 DPC. All animal procedures and care were conducted under the supervision and regulations of the Iowa State University Institutional Animal Care and Use Committee. Sample collection. Pigs were bled up to seven times, depending on their time of necropsy at 0, 14, 28, 42, 56, 70 and 96 DPC. Bronchial alveolar lavage fluid (BALF) for M. hyopneumoniae PCR analysis was collected from each pig at necropsy using 25 ml of 56

phosphate-buffered saline (PBS), pH 7.2 (1.54mM KH2PO4, 155.17mM NaCl, and 2.71mM

Na2HPO4-7H20) as previously described (13). Two bronchial swabs were collected from all pigs. In addition, two joint swabs from the right hock and left knee joints were collected from pigs in the M. hyosynoviae and M. hyorhinis groups. One swab from each sample-type was used for routine bacteriological analysis and the other was placed in 1 ml of PBS for PCR analysis. Swabs in PBS and BALF were stored at -20ºC prior to DNA extraction. Macroscopic pneumonia. At necropsy, lung lesions consistent with M. hyopneumoniae- induced pneumonia were sketched for each pig onto a standard lung diagram, then analyzed for the proportion of affected lung using a Zeiss SEM-IPS image analysis system as previously described (16). ELISA assays. Serology was performed on all serum samples with the IDEXX and Oxoid ELISAs performed as per manufacturer’s instructions and the 232 Tween 20 ELISA performed as previously described (4). To determine if the competitive-inhibition ELISA manufactured by Oxoid provided the same results as the previous manufacturer (DAKO Corporation, Carpenteria, Calif.), all sera were also tested by the DAKO M. hyopneumoniae ELISA. In the Oxoid and DAKO ELISAs, a sample is defined as positive if the percent inhibition of the sample is ≤ 50% of the buffer control. Samples with an OD > 50% of the buffer control are defined as negative. The IDEXX assay presents results as a S:P ratio which is calculated as: (sample OD – negative control OD) ÷ (positive control OD – negative control OD). A S:P ratio ≥ 0.4 is positive; < 0.4, ≥ 0.3 is suspect; and < 0.3 is negative. For the Tween 20 ELISA, all samples are tested in duplicate and the average optical density (OD) of the two is used. The positive control is normalized to an OD of 0.4, and the samples are then adjusted accordingly. Optical densities ≥ 0.24 are defined as positive. Negative results are defined as OD < 0.2, and an OD < 0.24, ≥ 0.2 is defined as suspect. In addition, seroconversion to M. flocculare, M. hyorhinis and M. hyosynoviae was evaluated using antigen made to each of these species by the Tween 20 method (4). Briefly, each mycoplasma was cultured to log-phase using the medium described above. Mycoplasma flocculare strain Ms42 ATCC 27399, M. hyorhinis strain 7, and M. hyosynoviae strain 97MS0003 cultures were harvested for 45 min at 15,000 xg, and then washed three times with PBS. Tween 20 was added at a final concentration of 1% to a 2 mg/ml mycoplasma cell suspension (BCA protein assay, Pierce, Rockford, Illinois). The mixtures were incubated at 37°C for 90 min then harvested at 50,000 xg. The supernatant was passed through a 0.2 µm filter. Antigen was coated at a concentration of 10 µg/ml (100 µl/well) in 96-well Immulon 2HB plates (Thermo electron corp., Milford, MA) overnight at room temperature, then the plates were frozen at -80°C until ELISAs were performed following the same steps as for the 57

232 Tween 20 ELISA (4). In addition, the Tween 20 method was used to make ELISA plates from M. hyopneumoniae isolate 97MP0001. PCR from nasal swabs, bronchial swabs and bronchial alveolar lavage fluid. The QIAamp DNA mini kit (Qiagen, Valencia, Calif.) was used for DNA isolation from nasal swabs, bronchial swabs, joint swabs and BALF collected throughout the study. The BALF was processed as follows: a 1 ml aliquot from each pig was centrifuged at 16,000 x g for 10 min. The resulting pellet was resuspended in 200 µl of PBS, and then extracted as per manufacturer’s instructions. Mycoplasma hyopneumoniae was assayed using a real-time PCR targeting the mhp183 gene of M. hyopneumoniae as previously described (Strait et al., submitted). Western blot assay. A western blot was performed using the Tween 20 ELISA antigen prepared for isolate 97MP0001. The antigen was treated in Laemmli sodium dodecyl sulfate- polyacrylamide gel electrophoresis (SDS-PAGE) reducing buffer. The SDS-PAGE method by Laemmli (11) was followed using a 4% stacking and 10% separating gel. A total of 600 µl (300 µg/ml) of antigen was loaded in one large well spanning most of the gel. Separated proteins were transferred to a nitrocellulose membrane (Bio-Rad, Hercules, Calif.). The membrane was blocked overnight at 4°C in PBS with 1% gelatin from cold water fish skin (Sigma, St. Louis, Missouri), then placed in a miniblotter® 25 (Immunetics, Cambridge, Mass.) to form individual lanes. To each lane, 250 µl of a 1:40 dilution of serum in PBS with 0.05% Tween 20 (PBS-T) was added and incubated for 1.5 hr, then washed 3 times with PBS-T for 10 min. Serum samples at 0 DPC and 56 DPC from pigs in each of the M. hyopneumoniae challenge groups were used (see Figure 4.6). Horseradish peroxidase-labeled rabbit anti-swine (MP Biomedicals, Aurora, Ohio) diluted 1:1000 in PBS-T was added and incubated for 1 hr, then washed as above. The substrate 4CN (Bio-Rad) was used as per manufacturer’s instructions. Statistics. The ODs within each ELISA were evaluated by an analysis of variation (ANOVA) to determine if there were differences between mean ODs of the groups. Fisher’s least significant difference (LSD) was used to compare group means at each time point when the ANOVA was significant at a level of p < 0.05.

Results Macroscopic pneumonia. Pigs in all of the M. hyopneumoniae challenge groups had lung lesions consistent with mycoplasmal pneumonia, primarily at the 28 DPC necropsy (Table 4.1). Pigs in groups challenged with M. flocculare, M. hyorhinis or M. hyosynoviae had negligible macroscopic pneumonia throughout the study (Table 4.2). 58

Bacteriological findings. Culture results are summarized in Tables 4.1 and 4.2. M. hyopneumoniae was isolated from all pigs in each of the four M. hyopneumoniae challenge groups at 28 DPC, with fewer M. hyopneumoniae isolations at either 70 or 96 DPC. Additionally, M. flocculare was cultured from 5 of 6 pigs in the 00MP1502 M. hyopneumoniae challenge group. Mycoplasma flocculare was re-isolated from the two M. flocculare groups. In the M. hyosynoviae challenge group M. hyosynoviae was re-isolated from one pig, and M. flocculare was found in 5 of 6 pigs in this group. Mycoplasma hyorhinis was re-isolated from four pigs in the M. hyorhinis challenge group and these same four pigs also had M. hyopneumoniae cultured from them. No other pathogenic bacteria were cultured from any pig. Oxoid, IDEXX and 232 Tween 20 ELISAs. The Oxoid, IDEXX and 232 Tween 20 ELISAs detected antibodies in pigs from all four of the M. hyopneumoniae challenge groups (Figures 4.1, Table 4.3). All pigs were seronegative for M. hyopneumoniae antibodies prior to challenge. Seroconversion occurred as early as 14 DPC in 22 of 24 pigs as measured by the Oxoid ELISA, none of the M. hyopneumoniae-infected pigs were seropositive at that time-point using the IDEXX or 232 Tween 20 ELISAs. By 28 DPC all pigs in all challenge groups were seropositive using the Oxoid ELISA and remained positive until necropsy. Alternatively at the 28 DPC time-point, the 232 Tween 20 ELISA detected antibody positive or suspect values in 11 of 23 (48%) of the M. hyopneumoniae-infected pigs, while the IDEXX ELISA detected antibodies in 16 of 23 (70%) pigs. By 42 DPC both indirect ELISAs detected M. hyopneumoniae antibodies in 14 of 15 (93%) of the M. hyopneumoniae-infected pigs. Among pigs that were kept longer than 28 DPC, all were found positive at one time point by the Oxoid and IDEXX ELISAs, while pig 157, challenged with M. hyopneumoniae isolate 97MP0001, remained seronegative at all time-points by the 232 Tween 20 ELISA (Table 4.3). There was no difference in average optical density (OD) observed among any of the groups by the Oxoid assay (data not shown), but one M. hyopneumoniae isolate (97MP0001) induced significantly lower antibody levels compared to the three other M. hyopneumoniae challenge groups as measured by OD using 232 Tween 20 ELISA (Figures 4.1) and also the IDEXX ELISA (data not shown).

Table 4.1. Summary of the percentage of pneumonia, culture, and PCR results for pigs challenged with M. hyopneumoniae field isolates. PCR results from NS, BS and BAL for each pig versus DPC Culture results 7 14 21 28 42 56 70 96 Necropsy Pneumonia Group (DPC) ID (%) NS BS NS NS NS NS BS BAL NS NS NS BS BAL NS BS BAL 169 16.18 Mhp - + + + + + 28 172 22.31 Mhp - - + + + + 153 23.79 + - + + + - - + + 70 157 0.00 + + + + + - - - + Mhp 163 0.12 + + + + + - - - - - 97MP0001 96 170 1.21 Mhp + - + + - + - - + - 158 10.62 Mhp - + + + + + 28 162 31.52 Mhp + + + + + + 159 0.00 + + + + + + + - - 70 160 0.01 + + + + + + + + +

Mhp 161 0.82 Mhp + + + + + + + + + + 59 95MP1501 96 164 1.73 Mhp + + + + + + - - + + 755 12.98 Mhp + - + + + + 28 762 2.32 Mhp + + + - + + 70 773 0.00 + + - + - - - + + Mhp 752 0.48 + + - + - - + - - - 00MP1301 96 767 0.15 - + + + - - + - - - 768 27.56 Mhp, Mfl Mhp - + + + + + 28 772 16.59 Mhp, Mfl Mhp + + + + + + 757 6.94 Mhp - + + - + - + + + 70 759 21.79 Mfl Mhp + - + + - - + + + Mhp 751 0.83 Mfl + - - - - + - - + + 00MP1502 96 770 0.21 Mhp, Mfl + + + + + + + + + + DPC = days-post-challenge; ID = pig identification number; NS = nasal swab; BS = bronchial swab; BAL = bronchial alveolar lavage fluid; Mhp = M. hyopneumoniae; Mfl = M.flocculare. 60

Table 4.2. Summary of pneumonia and culture results for pigs challenged with M. flocculare, M. hyorhinis and M. hyosynoviae. Group Necropsy (DPC) ID Pneumonia (%) Culture results NS BS 156 0.00 Mfl Mfl 28 166 0.57 Mfl 154 0.41 Mfl 70 173 0.19 Mfl Mfl 167 0.14 Mfl Mfl 7161B 96 168 0.12 174 0.30 Mfl Mfl 28 175 2.18 Mfl Mfl 70 171 0.28 Mfl Mfl 152 0.36 Mfl 48LB 96 155 0.11 Mfl 758 0.22 Mhr, Mhp Mhr, Mhp 28 769 0.07 Mhr Mhr, Mhp 754 0.58 Mhr Mhr, Mhp 70 763 0.04 Mhr Mhr, Mhp 756 0.32 Mhr 96 774 0.31 753 0.34 Mfl Mfl 28 760 0.00 Mhs, Mfl Mfl 761 0.17 70 766 0.92 Mfl 765 0.11 Mfl Mhs 96 771 0.05 Mfl DPC = days-post-challenge; Mfl = M. flocculare; Mhr = M. hyorhinis; Mhp = M. hyopneumoniae; Mhs = M. hyosynoviae; ID = pig identification number; NS = nasal swab; BS = bronchial swab.

The Oxoid competitive-inhibition ELISA was previously manufactured as the DAKO M. hyopneumoniae ELISA. In order to compare the kits manufactured by Oxoid and DAKO, all sera were assayed by both ELISAs. No difference was observed in results produced by sera tested by either the DAKO, or the Oxoid ELISAs (data not shown). 61

Figure 4.1. 232 Tween 20 OD comparison of group means in M. hyopneumoniae-infected groups versus days-post-challenge. At each day-post-challenge, data points labeled with different lower case numbers letters are significantly different from each other, p < 0.05. Points above the horizontal line are considered positive.

All pigs in the M. flocculare, M. hyorhinis and M. hyosynoviae groups seroconverted to the mycoplasma with which they were challenged. This was determined by observing rising OD values over time in the ELISAs performed for each species of mycoplasma (data not shown). In addition, pigs in the M. hyosynoviae group and the 00MP1502 M. hyopneumoniae group, seroconverted to M. flocculare. The Oxoid, IDEXX and 232 Tween 20 ELISAs detected no antibodies from any of the pigs in groups experimentally challenged with M. flocculare, M. hyorhinis or M. hyosynoviae. Although M. hyopneumoniae was cultured from pigs in the M. hyorhinis group; these pigs did not develop detectable antibodies to M. hyopneumoniae.

Table 4.3. ELISA results from pigs challenged with M. hyopneumoniae field isolates. Oxoid, IDEXX, 232 Tween 20 ELISA Results* Days post challenge (DPC) Group ID 0 14 28 42 56 70 96 97MP0001 153 -, -, - -, -, - +, -, - +, +, + +, +, + +, +, + 157 -, -, - +, -, - +, +, - +, +, - +, +/-, - +, -, - 163 -, -, - +, -, - +, +/-, +/- +, -, + +, +, + +, +, +/- +, +, +/- 169 -, -, - +, -, - +, -, - 170 -, -, - +, -, - +, +/-, - +, +, + +, -, + +, -, +/- +, -, +/- 172 -, -, - +, -, - +, -, - 95MP1501 158 -, -, - +, -, - +, +, + 159 -, -, - +, -, - +, +/-, + +, +, + +, +, + +, +, + 160 -, -, - +, -, - +, -, - +, +, + +, +, + +, +, + 161 -, -, - +, -, - +, +/-, - +, +, + +, +, + +, +, + +, +, + 162 -, -, - +, -, - +, +/-, - 164 -, -, - +, -, - +, -, - +, +, + +, +, + +, +, + +, +, + 00MP1301 752 -, -, - +, -, - +, +/-, + +, +, + +, +, + +, +, + +, +, + 755 -, -, - +, -, - +, -, - 762 -, -, - +, -, - +, +/-, - 767 -, -, - +, -, - +, +, + +, +, + +, +, + +, +, + +, +, + 773 -, -, - +, -, - +, +/-, +/- +, +, + +, +, + +, -, + 00MP1502 751 -, -, - +, -, - +, +/-, + +, +, + +, +, + +, +, + +, +, + 757 -, -, - +, -, - +, +/-, - +, +, + +, +, + +, +, + 759 -, -, - +/-, -, - +, -, +/- +, +, + +, +, + +, +, + 768 -, -, - +, -, - +, +, + 770 -, -, - +, -, - +, +, + +, +, + +, +, + +, +, + +, +, + 772 -, -, - +, -, - +, +/-, +/- * Samples are categorized as positive (+), negative (-), or suspect (+/-) by the Tween 20 and IDEXX ELISAs and + or – by the Oxoid ELISA. ID = pig identification number.

97MP0001 Tween 20 ELISA. Serum from all pigs in each of the four M. hyopneumoniae challenge groups were tested using the 97MP0001 Tween 20 ELISA plates. The results are summarized as group averages for each time point in Figure 4.2. The only time-point when any groups were significantly different from each other was 42 DPC.

Figure 4.2. 97MP0001 Tween 20 OD comparison of group means in M. hyopneumoniae- infected groups versus days-post-challenge. At 42 days-post-challenge, data points labeled with different lower case numbers letters are significantly different from each other, p < 0.05.

PCR. Results from the M. hyopneumoniae PCR assays on nasal and bronchial swabs and BALF are summarized in Table 4.1. More positive results were obtained at 28 DPC than at either 70, or 96 DPC. Samples from BALF and bronchial swabs were each positive in 21 of 23 samples while samples collected from nasal swabs were positive in 100 of 145 samples. The number of positive results for each sample-type peaked at 28 DPC. Western blot. Sera from pigs in each of the M. hyopneumoniae challenge groups were evaluated in a western blot assay using Tween 20-extracted antigen from isolate 97MP0001 (Figure 4.3). Except for sera from pig 153, the intensity of bands from pigs challenged with isolate 97MP0001 were less than was observed from pigs challenged with the other three M. hyopneumoniae isolates. Pig 153 was the only pig in the 97MP0001 challenge-group that was positive by all three ELISAs at the 40, 56 and 70 DPC time-points.

Figure 4.3. Western blot results detecting proteins from antigen prepared from M. hyopneumoniae isolate 97MP0001. Included are sera from pigs in each of the four M. hyopneumoniae challenge-groups at 0 and 56 days-post-challenge (DPC). Numbers along the left side of the figure indicate molecular weight markers. ID = Pig identification number.

Discussion Enzyme-linked immunosorbent assays are the most commonly used tool to monitor herds and screen replacement animals for exposure to M. hyopneumoniae. Here the sensitivities and specificities of three M. hyopneumoniae ELISA assays used in the United States were compared. Groups of pigs were experimentally infected with four species of mycoplasmas commonly found in swine. In addition, multiple isolates of M. hyopneumoniae and M. flocculare were included to further examine the effect of strain variation within species. The current study differs from previous studies directly comparing the sensitivity and specificity of the Tween 20, IDEXX and Oxoid (Dako) M. hyopneumoniae-specific ELISAs which used sera collected from pigs challenged with M. hyopneumoniae strain 232 and/or field samples (1, 7) and had no groups experimentally challenged with other Mycoplasma species. Strain 232 is used to produce the antigen for the Tween 20 ELISA and was derived from an in vivo pig passage of strain 11, which is one of the original strains of M. hyopneumoniae (23). Testing sera collected from pigs 65 challenged with strain 232 may not be representative of current conditions in the field. A number of other species of mycoplasmas and acholeplasmas have also been isolated from pigs (22), but their prevalence is unclear. The effect antibodies to these organisms have on the specificity of M. hyopneumoniae ELISAs was not examined, but may warrant further study. In this study the Oxoid ELISA was the most sensitive assay detecting antibodies in pigs at 14 DPC, the earliest time point that sera was collected after infection. Neither the 232 Tween 20, nor the IDEXX ELISA detected antibodies until the 28 DPC time-point. The 232 Tween 20 and IDEXX ELISAs performed similarly to each other in this study overall, but differed 24% of the time on individual results from 28-96 DPC (Table 4.3). Within these differences there was an equal detection of positive, suspect, or negative by each ELISA, rather than one assay being consistently more positive than the other. Erlandson, et al. suggested that the positive-cut off in the 232 Tween 20 ELISA was set more conservatively than either the IDEXX or DAKO (now Oxoid) assays (7). Although in the current study lowering the cut-off of the 232 Tween 20 ELISA did not substantially improve the agreement between the two indirect ELISAs (data not shown). Higher agreement between the two indirect assays might be expected because the antigens used in the IDEXX and 232 Tween 20 ELISAs are produced by the same technique with the primary difference between the tests being the isolate of M. hyopneumoniae used for antigen production. Unlike the immune response to many pathogens, antibodies to M. hyopneumoniae can take weeks to months to reach detectable levels after infection. Under field conditions challenge dose may further impact the rate of seroconversion; therefore, multiple time- points for serology may be necessary to identify positive pigs. Combining the Oxoid ELISA with either the 232 Tween 20 or IDEXX ELISA increases sensitivity as has been previously reported (7). Few reports of false-positives in any of these ELISAs exist (4). In the current study, the M. flocculare, M. hyorhinis and M. hyosynoviae groups each had pigs with detectable antibodies to the species of mycoplasmas with which they were challenged. The M. hyorhinis challenge group was also cultured positively for M. hyopneumoniae, but these pigs did not develop detectable antibodies to M. hyopneumoniae by any of the ELISAs (data not shown). The M. hyosynoviae group and one were also culture-positive for M. flocculare. These groups did seroconvert to M. flocculare (data not shown). It was not determined whether these additional infections came from the source-herd, or were due to cross-contamination between challenge groups during the study. 66

A previous study compared the DAKO ELISA (now manufactured as the Oxoid ELISA) to the IDEXX ELISA and identified field sera that were positive by the DAKO ELISA, but negative by the IDEXX ELISA and vice versa (1). The discrepancies in the results between these tests were further assessed using an immunoblot assay (2) and it was suggested that a sample found positive by either test was truly positive, although this requires further validation as the true infection-status of the animals in this study was unknown. When the M. hyopneumoniae-specific ELISAs were compared for their ability to detect antibodies to different field isolates of M. hyopneumoniae, variation in the 232 Tween 20 and IDEXX ELISAs was observed. Specifically, pigs challenged with M. hyopneumoniae isolate 97MP0001 had significantly lower ODs than was observed in any of the other M. hyopneumoniae groups. One explanation for the lower ODs observed in the 97MP0001 group is that pigs challenged with the 97MP0001 isolate had a less severe challenge; however, the percentage of macroscopic pneumonia indicates otherwise. At 28 DPC, the time-point normally associated with the most severe pneumonia after experimental challenge, pigs in the 00MP1301 group appeared to have the least amount of pneumonia. In addition, M. hyopneumoniae was detected in pigs from all four M. hyopneumoniae challenge groups until the end of the study at 96 DPC by PCR indicating that isolate 97MP0001 was not cleared from the pigs at a different rate than the other three M. hyopneumoniae isolates. In this study, each time point was represented by only two pigs which does not allow for statistical comparisons to be made. Another possible explanation for the lower OD levels is that antibodies produced in the pigs challenged with isolate 97MP0001 less effectively bound the antigen used in the 232 Tween 20 and IDEXX ELISAs. To test this, sera from pigs in all of the M. hyopneumoniae groups were assayed using a Tween 20 ELISA with antigen produced from isolate 97MP0001. Similar to the results observed with the IDEXX ELISA and the 232 Tween 20 ELISA, pigs challenged with isolate 97MP0001 had the lowest OD values in the 97MP0001 Tween 20 ELISA. However differences between the three indirect ELISAs did exist as the OD values observed for the four challenge groups in each of the ELISAs was not equivalent suggesting that at least some of the variation can be attributed to the antigen. A final explanation for the differences in results observed in sera from pigs in the 97MP0001 group may be due to lower levels of antibodies induced by the isolate. A western blot using sera from each of the M. hyopneumoniae groups before challenge (0DPC) and at the time of peak ODs (56 DPC) was performed using 97MP0001 Tween 20 antigen. Again even though, antibodies to homologous antigen might have been expected to produce the strongest bands, this was not observed. Except for pig 156, it 67 appears that fewer antibodies were produced by pigs challenged with isolate 97MP0001 than by pigs challenged with the other three M. hyopneumoniae isolates. Pigs in the M. hyopneumoniae group 00MP1502 were also infected with M. flocculare. It is unclear if antibodies to M. flocculare are being detected in the western blot, or what affect they may have on these results. Field isolate 97MP0001 could potentially possess a mechanism for skewing the immune response away from the humoral response. In the present study no differences were observed in the ability of the Oxoid ELISA to detect antibodies against any of the field isolates at any time-point, unlike what was observed in the Tween 20 and IDEXX ELISAs. The Tween 20 and IDEXX ELISAs are indirect assays using antigen prepared from Tween 20 extracts of whole-cell organisms. In these two assays, rising OD values directly correlate to increasing levels of antibodies. The Oxoid ELISA is a competitive-inhibition assay measuring antibodies to a single protein. In this assay, OD levels do not correlate in a linear manner to antibody levels. Distinguishing between samples with different levels of antibodies using the Oxoid ELISA is more difficult compared to the indirect ELISAs. This likely explains the differences observed between the two types of ELISAs. Bereiter, et al. found that pigs with high antibody levels against M. flocculare can potentially cause false-positive results in the 232 Tween 20 ELISA (4). In the current study sera from pigs challenged with M. flocculare had rising OD values in the M. flocculare Tween 20 ELISA over time and appeared to seroconvert. No evidence of cross-reactivity was attributed to the M. flocculare antibodies in the current study. However, peak M. flocculare OD values in these samples were 3-5 times lower than was observed by Bereiter, et al. As a result, the ability of high levels of M. flocculare antibodies to cross-react in the 232 Tween 20 ELISA remains unresolved. Pigs in the M. hyopneumoniae group, 00MP1502, and the M. hyosynoviae group were also infected with M. flocculare; however, these ODs were also lower than was observed by Bereiter, et al. The M. hyopneumoniae PCR results from nasal and bronchial swabs and BALF showed that early in infection all of these samples effectively produced positive results. At later necropsy time-points, BALF was more likely to have positive results followed by bronchial swabs, and then nasal swabs as has been previously shown (10). In addition, culture of M. hyopneumoniae was more likely to occur from pigs necropsied at 28 DPC than those necropsied at either 70 or 96 DPC and bronchial swabs were much more often successful than nasal swabs. In this study the swabs and BALF for PCR and culture were collected from pigs experimentally challenged at an infectious dose much higher than would occur in the field, but still demonstrate that the use of nasal swabs late in infection 68 is more likely to produce false-negative results than either BALF or bronchial swabs for both PCR and culture. The results of the current study suggest that seroconversion to field isolates of M. hyopneumoniae in pigs is not uniform. More studies need to be performed with the indirect ELISAs to determine if the M. hyopneumoniae isolates currently used for antigen production should be updated. Differences in virulence among isolates of M. hyopneumoniae did not appear to correlate to the level of antibodies detected by either ELISA or western blot, although this should be further investigated in a study using more pigs. It is currently unknown if antibodies to any proteins occur early in infection versus late, or what the effect of a low versus high infectious dose has on the detection rate of ELISAs. Therefore, M. hyopneumoniae serology is best used on a herd basis rather than on an individual pig basis, and screening of individual replacement animals by ELISA requires careful interpretation.

Acknowledgements This project was funded by a grant from the National Pork Board. The authors thank Nancy Upchurch, and the students in the Thacker laboratory for technical assistance.

References

1. Ameri-Mahabadi, M., E. M. Zhou, and W. H. Hsu. 2005. Comparison of two swine Mycoplasma hyopneumoniae enzyme-linked immunosorbent assays for detection of antibodies from vaccinated pigs and field serum samples. J. Vet. Diagn. Invest. 17:61-64. 2. Ameri, M., E. M. Zhou, and W. H. Hsu. 2006. Western blot immunoassay as a confirmatory test for the presence of anti-Mycoplasma hyopneumoniae antibodies in swine serum. J. Vet. Diagn. Invest. 18:198-201. 3. Assuncao, P., C. De la Fe, A. S. Ramirez, O. Gonzalez Llamazares, and J. B. Poveda. 2005. Protein and antigenic variability among Mycoplasma hyopneumoniae strains by SDS-PAGE and immunoblot. Vet. Res. Commun. 29:563-74. 4. Bereiter, M., T. F. Young, H. S. Joo, and R. F. Ross. 1990. Evaluation of the ELISA and comparison to the complement fixation test and radial immunodiffusion enzyme assay for detection of antibodies against Mycoplasma hyopneumoniae in swine serum. Vet. Microbiol. 25:177-192. 69

5. Burnett, T. A., K. Dinkla, M. Rohde, G. S. Chhatwal, M. Srivastava, S. J. Cordwell, S. Geary, F. C. Minion, M. J. Walker, and S. P. Djordjevic. 2006. P159 is a proteolytically processed, surface adhesin of Mycoplasma hyopneumoniae: defined domains of P159 bind heparin and promote adherence to eukaryotic cells. Mol. Microbiol. 60:669-686. 6. Djordjevic, S. P., S. J. Cordwell, M. A. Djordjevic, J. Wilton, and F. C. Minion. 2004. Proteolytic processing of the Mycoplasma hyopneumoniae cilium adhesin. Infect. Immun. 72:2791-2802. 7. Erlandson, K. R., R. B. Evans, B. J. Thacker, M. W. Wegner, and E. L. Thacker. 2005. Evaluation of three serum antibody enzyme-linked immunosorbent assays for Mycoplasma hyopneumoniae. J. Swine Health Prod. 13:198-203. 8. Feld, N. C., P. Qvist, P. Ahrens, N. F. Friis, and A. Meyling. 1992. A monoclonal blocking ELISA detecting serum antibodies to Mycoplasma hyopneumoniae. Vet Microbiol 30:35-46. 9. Friis, N. F. 1975. Some recommendations concerning primary isolation of Mycoplasma suipneumoniae and Mycoplasma flocculare a survey. Nord. Vet. Med. 27:337-339. 10. Kurth, K. T., T. Hsu, E. R. Snook, E. L. Thacker, B. J. Thacker, and F. C. Minion. 2002. Use of a Mycoplasma hyopneumoniae nested polymerase chain reaction test to determine the optimal sampling sites in swine. J. Vet. Diagn. Invest. 14:463-469. 11. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685. 12. Madsen, M. L., M. J. Oneal, S. W. Gardner, E. L. Strait, D. Nettleton, E. L. Thacker, and F. C. Minion. 2007. Array-based genomic comparative hybridization analysis of field strains of Mycoplasma hyopneumoniae. J. Bacteriol. 189:7977-7982. 13. Mengeling, W. L., K. M. Lager, and A. C. Vorwald. 1995. Diagnosis of porcine reproductive and respiratory syndrome. J Vet Diagn Invest 7:3-16. 14. Nicolet, J., P. Paroz, and S. Bruggmann. 1980. Tween 20 soluble proteins of Mycoplasma hyopneumoniae as antigen for an enzyme linked immunosorbent assay. Res. Vet. Sci. 29:305-309. 15. Ross, R. F., and J. A. Karmon. 1970. Heterogeneity among strains of Mycoplasma granularum and identification of Mycoplasma hyosynoviae, sp. n. J. Bacteriol. 103:707-713. 70

16. Ross, R. F., B. J. Zimmermann-Erickson, and T. F. Young. 1984. Characteristics of protective activity of Mycoplasma hyopneumoniae vaccine. Am. J. Vet. Res. 45:1899-1905. 17. Scarman, A. L., J. C. Chin, G. J. Eamens, S. F. Delaney, and S. P. Djordjevic. 1997. Identification of novel species-specific antigens of Mycoplasma hyopneumoniae by preparative SDS-PAGE ELISA profiling. Microbiology 143:663-673. 18. Sorensen, V., P. Ahrens, K. Barfod, A. A. Feenstra, N. C. Feld, N. F. Friis, V. Bille-Hansen, N. E. Jensen, and M. W. Pedersen. 1997. Mycoplasma hyopneumoniae infection in pigs: duration of the disease and evaluation of four diagnostic assays. Vet. Microbiol. 54:23-34. 19. Sorensen, V., K. Barfod, and N. C. Feld. 1992. Evaluation of a monoclonal blocking ELISA and IHA for antibodies to Mycoplasma hyopneumoniae in SPF- pig herds. Vet. Rec. 130:488-490. 20. Sorensen, V., K. Barfod, N. C. Feld, and L. Vraa-Andersen. 1993. Application of enzyme-linked immunosorbent assay for the surveillance of Mycoplasma hyopneumoniae infection in pigs. Rev. Sci. Tech. 12:593-604. 21. Strait, E., B. Erickson, and E. Thacker. 2004. Presented at the Proc AASV, Des Moines, IA. 22. Thacker, E. L. 2006. Mycoplasmal diseases, p. 701-717. In B. E. Straw, Zimmerman, J.J., D'Allaire, S., Taylor, D.J. (ed.), Diseases of Swine, 9 ed. Blackwell, Ames. 23. Zielinski, G. C., and R. F. Ross. 1990. Effect of growth in cell cultures and strain on virulence of Mycoplasma hyopneumoniae for swine. Am. J. Vet. Res. 51:344-348. 71

A paper accepted to the Journal of Swine Health and Production

Erin L. Strait, DVM; Vicki J. Rapp-Gabrielson, MS, PhD; Barbara Z. Erickson, MS; Richard B. Evans, PhD; Lucas P. Taylor, MS; Therese K. Yonkers, BS; Rob L. Keich, MS; Rika Jolie, DVM, PhD, MBA; Eileen L. Thacker, DVM, PhD, Diplomate ACVM

ELS, BZE, RBE, ELT: Department of Veterinary Microbiology and Preventive Medicine, College of Veterinary Medicine, Iowa State University, Ames, IA 50011.

VJRG, LPT, TKY, RLK, RJ: Pfizer Animal Health, 7000 Portage Rd, Kalamazoo, MI 49001.

Summary Objective: The efficacy of a Mycoplasma hyopneumoniae (M hyo) bacterin was evaluated against two contemporary field isolates of M hyo. Methods: Two challenge studies were performed in which pigs received two doses of saline (nonvaccinated groups), or a M hyo bacterin (vaccinated groups), followed three weeks later by intratracheal inoculation with one of two M hyo field isolates. Necropsies were performed 28 or 30 days-post-challenge (DPC). Vaccine efficacy was determined by evaluating macroscopic lung lesions, DNA levels of M hyo in bronchial alveolar lavage fluids (BALF), M hyo-specific immunoglobulin (Ig)A and IgG antibodies, and serum antibodies measured by the Tween 20 and DAKO ELISAs. Results: The percentage of macroscopic lung lesions and concentration of M hyo DNA in BALF samples following challenge with either isolate were significantly less in vaccinated

compared to nonvaccinated pigs. M hyo-specific mucosal IgA and IgG antibody levels in BALF were significantly higher in the vaccinated pigs compared to the nonvaccinated pigs. Implications: Under the conditions of this study, pigs vaccinated with a M hyo bacterin prior to challenge with virulent field isolates of M hyo had fewer M hyo organisms and less pneumonia compared to nonvaccinated pigs. Vaccination is an effective tool to reduce pneumonia induced by M hyo, although more studies need to be performed to determine protection against a wide range of field isolates.

Introduction Mycoplasma hyopneumoniae (M hyo) is the causative agent of enzootic pneumonia, but more importantly, it is a primary component of the porcine respiratory disease complex. The impact of disease caused by M hyo leads to significant economic losses in the swine industry today. M hyo is an extracellular pathogen that resides primarily on the ciliated respiratory epithelium of the lower respiratory tract where it induces a delayed humoral immune response in the pig. Currently available commercial bacterins are unable to prevent M hyo colonization in the lungs, but have been shown to significantly reduce the severity of pneumonia in vaccinated pigs.1,2 Most experimental studies evaluating the efficacy of M hyo bacterins have relied on challenge models with older reference strains of M hyo. Many of these studies have used a model developed at Iowa State University that consists of an intratracheal inoculation of a diluted lung homogenate containing pig-passaged strain 2321,2, that derives from strain 11 which was cultured in the 1960’s.3 Recently, a range of virulence among M hyo isolates has been demonstrated.4,5,6 In addition genotypic and phenotypic variation among isolates has been shown to exist, although specific implications of these differences have not yet been identified.7,8,9,10,11,12 It is currently unknown if differences in virulence factors or other putative immunogens affect the ability of current vaccines to protect against different field isolates of M hyo. As a result, there is a need to measure the efficacy of M hyo vaccines against a variety of field isolates. In the study reported here, pigs were assigned to one of two treatment groups that received either a M hyo bacterin or saline. An equal number of pigs from each treatment group were then challenged with one of two different contemporary M hyo field isolates. The M hyo isolates used in this study, 95MP1509 and 00MP1301,6 are genetically distinct from each other and from strain 232.12 The parameters used to measure efficacy included macroscopic lung lesions, M hyo level in the lungs as assessed by a real-time polymerase chain reaction

(PCR), M hyo-specific Immunoglobulin (Ig)G and IgA antibodies in bronchial alveolar lavage fluids (BALF) measured by enzyme-linked immunosorbent assays (ELISA), and M hyo-specific serum antibodies were assessed by Tween 20 and DAKO ELISAs. Materials and Methods Study animals Ninety-three weaned and castrated male crossbred pigs, 9 to 14 days of age from a commercial farrowing unit were transported to the Iowa State University Animal Research Facilities in Ames, Iowa. The source herd had no history of vaccination against M hyo and was serologically free of M hyo and porcine reproductive and respiratory virus. No clinical signs of swine influenza virus were present in the pigs at the time of arrival. All animal procedures and care were conducted under the supervision and regulations of the Iowa State University Institutional Animal Care and Use Committee. Experimental design Eighty-four pigs were distributed into two separately housed challenge groups. Each group was randomly assigned to 7 pens of 6 pigs according to a generalized block design. Within each pen, three pigs were vaccinated with the M hyo bacterin and three pigs received saline (nonvaccinates). Two groups of three procedural negative control pigs were separately housed in a different facility and received no treatment or challenge. Food and water were supplied ad libitum throughout the study. At approximately 3 and 5 weeks of age, pigs were injected intramuscularly (IM) in the right and left neck, respectively, with 2 mL of an M hyo bacterin (RespiSure®; Pfizer Animal Health, New York, New York) or 2 mL of 0.9% saline as per treatment assignment and designated as vaccinated and nonvaccinated pigs, respectively. At approximately 8 weeks of age, each challenge group of pigs was inoculated intratracheally with a M hyo broth culture containing between 107 and 108 color changing units (CCU) per mL of either field isolate 95MP1509 or 00MP1301 on 3 consecutive days designated 0, 1 and 2 days post challenge (DPC) as previously described.1 Both 95MP1509 and 00MP1301 were isolated from pigs submitted to the Iowa State University Veterinary Diagnostic Laboratory in 1995 and 2000, respectively. Due to the large number of pigs in the study, pigs challenged with isolate 95MP1509 were necropsied at 28 days DPC, and pigs challenged with 00MP1301were necropsied at 30 DPC. The investigators were blinded to vaccination status of the pigs until all the results from all study parameters were collected.

Serum antibody tests Blood for serology was collected prior to each vaccination, prior to challenge, and at necropsy. All serum samples were frozen at -20°C until the end of the study and then batch- tested by the Tween 20 ELISA13 and the DAKO M hyo blocking ELISA (DAKO Corporation, Carpenteria, California). For the Tween 20 ELISA, all samples were tested in duplicate and the average optical density (OD) of the two was used. The positive control was normalized to an OD of 0.4, and the samples were then adjusted accordingly. Optical densities ≥ 0.24 were defined as positive. Negative results were defined as OD < 0.2, and an OD < 0.24, ≥ 0.2 was defined as suspect. For the DAKO ELISA, a sample was defined as positive if the percent inhibition of the sample was ≤ 50% of the buffer control. Samples with an OD > 50% of the buffer control were defined as negative. Necropsy All animals, including the procedural negative-control pigs, were euthanized using an AVMA-approved euthanasia solution (Fatal-Plus; Vortech Pharmaceuticals, Ltd, Dearborn, Michigan) followed by exsanguination. Lung lesions consistent with M hyo pneumonia were sketched on a standard diagram and assessed for the proportion of lung surface with lesions using a Zeiss SEM-IPS image analyzing system (Carl Zeiss, Inc, Thornwood, New York) as previously described.1 Bronchial swabs were cultured for respiratory pathogens of swine including Bordetella bronchiseptica, Pasteurella multocida, Actinobacillus spp., and Haemophilus spp. Samples were inoculated onto blood agar plates and streaked with a Staphylococcus epidermidis nurse colony for support of A. pleuropneumoniae and Haemophilius spp. In addition, bronchial swabs were inoculated into BHI-TS mycoplasma medium14 for M. hyorhinis isolation and Friis medium15 for isolation of M hyo. Species identification of mycoplasma cultures was confirmed using a multiplex PCR for M hyo, Mycoplasma flocculare and Mycoplasma hyorhinis as previously described.16 BALF was collected using 25 mL of sterile phosphate buffered saline (PBS) as previously described.17 BALF samples were assayed by ELISA for local IgA and IgG antibodies to M hyo18 and for M hyo DNA by real-time PCR. For the real-time PCR, a 1 mL aliquot of BALF from each pig was centrifuged at 16,000g for 10 minutes. The resulting pellet was resuspended in 200 µL of PBS. DNA was isolated using the QIAamp DNA mini kit (Qiagen, Valencia, California). DNA was also isolated from M hyo strain 232 by phenol-chloroform

extraction19 for use in generating a standard curve for semi-quantitative analysis. The real- time PCR (REP assay), which targets the repeated element MHYP1-03-950 (accession no. AF004388), was performed with minor modifications as previously described.20 The concentration used for the probe was 250 nM. An internal positive control was added to control potential inhibition caused by the DNA sample (Applied Biosystems, Foster City, California). The 10X exogenous internal positive control mix and 50X exogenous internal positive-control DNA were diluted two-fold further than recommended by the manufacturer to prevent interference within the assay. All PCR reactions were performed in duplicate on a Rotor-Gene RG-3000 (Corbett Research, San Francisco, California). Results were quantified by comparison to a standard curve generated for each run using Rotor-Gene 5 software (Corbett Research) and expressed as ng per µL DNA, duplicate samples were then averaged together. Data analysis The two challenge groups were treated as separate studies for the purposes of randomization and statistical analysis. In all analyses, the individual pig was the experimental unit. The arcsine square root transformation was applied to the total percentage lung involvement prior to analysis, and the transformed percent lung scores were analyzed with a general linear mixed model with fixed term treatment and random term block. For each antibody test procedure, OD values were back-transformed to obtain geometric mean titers prior to analysis. Transformed serum antibody values were analyzed with a general linear repeated measures mixed model with fixed effects: treatment, time point, and treatment by time-point interaction, and random effects: block and animal within block by treatment. The positive-negative status for both Tween 20 and DAKO ELISAs were analyzed with a generalized linear repeated measures mixed model (GLIMMIX) with fixed effects: treatment, time point, and treatment by time-point interaction, and random effects: block and animal within block by treatment. Suspect results as described above for the Tween 20 assay were considered negative for the analysis. M hyo IgA and IgG mucosal antibody titers and real-time PCR concentrations from BALF were analyzed with a generalized linear mixed model with fixed effect treatment and random effects terms block. If the model (via the GLIMMIX macro) did not converge when analyzing positive-negative status, then Fisher’s exact test was used to analyze that variable. All analyses were performed using SAS/STAT guide version 8 (SAS institute, Inc, Cary, North Carolina). The 5% level of significance (P < .05) was used to assess statistical differences.

Results Death of animals during the study One pig in the nonvaccinated group (95MP1509 challenge) and two pigs in the vaccinated group (00MP1301 challenge) died of causes not related to the study. Serology The OD values for all pigs were classified as negative for M hyo-specific antibodies in the Tween 20 and DAKO ELISAs at beginning of the study and prior to vaccination. The nonvaccinated pigs remained negative for M hyo antibodies prior to challenge and the procedural negative control pigs remained serologically negative at all time points throughout the trial (data not shown). The percentage of antibody positive pigs as measured by the Tween 20 and DAKO assays are summarized in Table 5.1. The OD levels measured by the Tween 20 assay were significantly higher (P < .001) in vaccinated pigs than in nonvaccinated pigs from both the 95MP1501 and 00MP1301 challenge groups prior to challenge and at necropsy (Table 5.2). Macroscopic lung lesions Analysis of the percentage of macroscopic lung lesions is summarized in Table 5.3. The mean percent lung involvement in vaccinated pigs was significantly lower than observed in nonvaccinated pigs (P <. 001) in both challenge studies [95MP1509 (5.1% compared to 19.3%) and 00MP1301 (1.2% compared to 7.4%)]. The distribution of lung lesions by severity is summarized in Table 5.4. The macroscopic lung lesion scores for the procedural negative control pigs averaged 0.36% for the 00MP1301 challenge study and 1.1% for the 95MP1509 study. M hyo real-time PCR Detection of M hyo DNA in BALF samples taken at necropsy is summarized in Table 5.5. Following challenge with either isolate, the level of M hyo DNA detected in vaccinated pigs was significantly lower (P < .001) than the level detected in the nonvaccinated pigs.

Challenge -33DPC/ -19DPC/ Treatment Test -1DPC* 28/30DPC† isolate Vacc 1 Vacc 2

Tween 20 95MP1509 Saline 0 0 0 60 DAKO 0 0 0 100 M hyo Tween 20 0 0 76 100 bacterin DAKO 0 5 100 100 00MP1301 Saline Tween 20 0 0 0 67 DAKO 0 0 0 100 M hyo Tween 20 0 0 74 100 bacterin DAKO 0 5 100 100 *For each challenge isolate, the percentage of MHYO vaccinates positive by Tween 20 and DAKO test was higher (P < .001) than saline controls. †For each challenge isolate, the percentage of MHYO vaccinates positive by Tween 20 assay was higher (P < .01) than saline controls. DPC=days post challenge, Vacc 1=day of first vaccination, Vacc 2=day of second vaccination.

For the Tween 20 ELISA, the positive control was normalized to an OD of 0.4, and the samples were then adjusted accordingly. Optical densities ≥ 0.24 were defined as positive. Negative results were defined as OD < 0.2, and an OD < 0.24, ≥ 0.2 was defined as suspect. Suspect titers were treated as negative for analysis. For the DAKO ELISA, a sample was defined as positive if the percent inhibition of the sample was ≤ 50% of the buffer control. Samples with an OD > 50% of the buffer control were defined as negative. The positive- negative status for both Tween 20 and DAKO ELISAs was analyzed with a generalized linear repeated measures mixed model (GLIMMIX) with fixed effects: treatment, time point, and treatment by time point interaction, and random effects: block and animal within block by treatment. If the model (via the GLIMMIX macro) did not converge when analyzing positive-negative status, then Fisher’s exact test was used to analyze that variable at each time point.

Table 5.2. M hyo Tween 20 serum antibody titers in pigs vaccinated twice with an M hyo bacterin, or saline, then challenged with M hyo, and necropsied 28 DPC (95MP1509 challenge) or 30DPC (00MP1301 challenge). No. of Tween 20 serum antibody titers (Mean OD ± SE)* Challenge Treatment pigs -33DPC/Vacc -19DPC/Vacc

isolate -1DPC 28/30DPC 1 2

20 0.031 + Saline 0.022 + 0.000 0.016 + 0.000 0.273 + 0.004 0.000 95MP1509 M hyo 21 0.323 + 0.018 + 0.000 0.021 + 0.000 0.962 + 0.017 bacterin 0.008

21 0.026 + Saline 0.025 + 0.000 0.017 + 0.000 0.292 + 0.006 0.000 00MP1301 M hyo 19 0.340 + 0.023 + 0.000 0.014 + 0.009 0.935 + 0.028 bacterin 0.009

* Back-transformed geometric mean titer ± standard error.

† M hyo bacterin titers were significantly higher (P < .001) than the saline placebo in both challenge groups at the -1DPC and 28/30DPC time points. DPC=days post challenge, Vacc 1=day of first vaccination, Vacc 2=day of second vaccination. Serum antibody titers were log transformed and analyzed with a general linear repeated measures mixed model with fixed effects treatment, time point, and treatment by time-point interaction, and random effects block and animal-within-block by treatment. The individual pig was the experimental unit.

IgA and IgG antibodies Significantly higher (P < .001) OD levels corresponding to M hyo-specific mucosal IgA and IgG antibodies were detected in vaccinated pigs compared to nonvaccinated pigs in both challenge groups (Table 5.6). The IgA and IgG optical density (OD) values for all procedural negative control pigs were low (OD < 0.07 for all samples; data not shown).

95MP1509 Saline 20 19.3 + 3.1 3 - 47

M hyo bacterin 21 5.1 + 1.7† 0 - 43

00MP1301 Saline 21 7.4 + 1.4 1- 28

M hyo bacterin 19 1.2 + 0.6‡ 0 - 19

* Back-transformed least squares mean + standard error. † M hyo vaccinates lower (P< .001) than saline controls. ‡ M hyo vaccinates lower (P< .001) than saline controls.

The arcsine square root transformation was applied to the total percentage lung involvement prior to analysis. The transformed percent lung scores were analyzed with a general linear mixed model with fixed term treatment and random term block

Bacterial isolation from bronchial swabs No primary swine bacterial pathogens were isolated from any pigs at necropsy. Streptococcus suis was isolated from nine and three pigs in the 95MP1509 and 00MP1301 challenge groups, respectively, but no clinical disease associated with this bacterium was observed. Culture for the detection of viable M hyo was positive in 19 of 20 nonvaccinated pigs and 18 of 21 vaccinated pigs challenged with 95MP1509 and for 20 of 21 nonvaccinated and 16 of 19 vaccinated pigs challenged with 00MP1301. There were no differences in the isolation rate for any bacterial species between any of the groups. All procedural negative control pigs from each challenge study were negative for isolation of all bacterial species except for one pig from which M hyo was cultured.

Table 5.4. Frequency distribution of percentage lung lesions at necropsy in pigs vaccinated twice with an M hyo bacterin, or saline, then challenged with M hyo, and necropsied 28 DPC (95MP1509 challenge) or 30DPC (00MP1301 challenge). Frequency Distribution of Lung Lesion Score No Challenge (percentage of pigs per range) Treatment of Isolate 5 < 10 < 20 < Pigs < 2% 2 < 5% > 30% 10% 20% 30%

Saline 20 0% 5% 20% 30% 20% 25%

95MP1509 MHYO 21 48% 24% 5% 5% 14% 5% bacterin

Saline 21 14% 29% 19% 29% 10% 0%

00MP1301 MHYO 19 68% 21% 0% 11% 0% 0% bacterin

95MP1509 Saline 2.415 + 0.622 0.49-11.29

M hyo bacterin 0.378 + 0.095† 0.02-2.62

00MP1301 Saline 1.188+ 0.502 0.01-6.14

M hyo bacterin 0.110 + 0.049‡ 0.00-2.32

*Back-transformed geometric mean titer + standard error. † M hyo vaccinates lower (P<0.001) than nonvaccinates. ‡ M hyo vaccinates lower (P<0.001) than nonvaccinates. A log transformation was applied to titer values and titers were analyzed with a generalized linear mixed model with fixed effect treatment and random effects block.

Challenge isolate Treatment IgA† IgG†

(GMT OD + SE*) (GMT OD + SE*)

Saline 0.236 + 0.017 0.377 + 0.025 95MP1509 M hyo bacterin 0.486 + 0.035 0.970 + 0.062

Saline 0.257 + 0.018 0.399 + 0.041 00MP1301 M hyo bacterin 0.458 + 0.020 0.932 + 0.041

*Back-transformed geometric mean titer optical density (OD) ± standard error. †For both challenge studies, immunoglobulin (Ig)A and IgG OD titers in the M hyo vaccinates were higher (P<0.001) than saline controls. For each antibody isotype a log transformation was applied to OD titer values and analyzed with a generalized linear mixed model with fixed effect treatment and random effects block. Discussion This study investigated the efficacy of a M hyo bacterin against two contemporary M hyo isolates. These isolates had previously been shown to produce more macroscopic lung lesions than the prototypic challenge strain 232.6 Following challenge with either strain of M hyo, the severity of macroscopic lung lesions in the vaccinated pigs was significantly less than in the nonvaccinated pigs. In both challenge studies, vaccinates had significantly lower concentrations of M hyo DNA in BALF, as detected by real time-PCR. In addition, higher OD values relating to mucosal M hyo-specific IgA and IgG antibodies were present in BALF samples collected at necropsy from vaccinated compared to nonvaccinated pigs. No differences were observed in the percentage of pigs from which M hyo was isolated between vaccinates and nonvaccinates in the two challenge groups at necropsy. One of the pigs in the procedural negative controls cultured positive for M hyo. This was likely due to post-harvest contamination as the procedural control pigs were housed in a different facility than the challenged pigs and did not exhibit any clinical signs of pneumonia at necropsy. A low percentage of macroscopic lesions was observed in the lungs of the some of the procedural

control pigs, but this is consistent with background levels normally observed in our challenge models using conventional pigs. Serology is a common tool used for monitoring herd status of M hyo as well as determining vaccination compliance. It has been suggested using field sera that the immune response in pigs infected with M hyo is not uniform and that these differences may impact the sensitivity of M hyo-specific ELISAs.21 Therefore two different M hyo ELISAs were used to evaluate M hyo serum antibodies in this study. Both the Tween 20 and DAKO ELISAs detected M hyo-specific serum antibodies in vaccinated pigs from each challenge group after two vaccinations. The antibody levels in the vaccinated pigs increased following challenge indicative of an anamnestic response. Direct comparisons of ODs between the vaccinates and nonvaccinates were made using Tween 20 results as OD values in this assay correlate to M hyo-specific antibody levels. In contrast, the DAKO ELISA is a competitive inhibition assay and OD values do not directly correlate to antibody levels. In the current study, the DAKO ELISA was more sensitive at earlier time-points after challenge with either field isolate similar to what has been previously shown using sera from pigs experimentally challenged with M hyo strain 232.22 Anecdotal reports from the field suggest that variation in M hyo isolates may play a role in mycoplasmal vaccine failure. Results from the studies presented here indicate that the M hyo vaccine provided protection against mycoplasmal pneumonia in pigs challenged with more recent M hyo field isolates. This challenge model is a more severe exposure than would occur in a natural field infection. Because challenge with culture may be less reproducible than with lung inoculum, pigs were inoculated on three consecutive days as has been previously described.23 In addition, both vaccinated and non-vaccinated pigs were co- mingled. Therefore, any effect of population immunity on shedding of M hyo was lost. Although these studies were analyzed separately and therefore were not directly compared, the 95MP1501 challenge resulted in more pigs in the saline treatment group with higher gross lesion scores (> 10%) than was observed in the saline treatment group in the 00MP1301 challenge. In addition, 19% of vaccinated pigs in the 95MP1501 group had pneumonia affecting > 20% of the lung, while most of the vaccinated pigs (89%) in the 00MP1301 group had lung lesions affecting < 5% of the lung. The severity of the challenge model may partially explain the high lung-lesion scores observed in pigs challenged with isolate 95MP1509. Alternatively this may demonstrate a lack of protection in these pigs against this isolate.

Currently, the importance of genetic variation on cross protection between M hyo isolates is not known. Since the virulence factors for M hyo have not been identified, the protective antigens required in a vaccine remain undefined. Neither of the field isolates used in this study has been used in commercial vaccine production. In addition, it is not known how genetically similar these two field isolates are to strains of M hyo currently used in M hyo bacterins. No M hyo genes, or gene products have been correlated to protection against disease caused by M hyo; therefore, no suitable targets for sequencing exist for comparison of field isolates in relation to virulence or protection from disease. This report measured vaccine efficacy following experimental challenge with two recent virulent M hyo field isolates. Further studies need to be performed to evaluate genetic and antigenic differences in M hyo isolates to better understand their implications in disease and vaccine efficacy. Implications Under the conditions of these studies, fewer M hyo organisms and less macroscopic lesions of pneumonia are observed when pigs are vaccinated with an M hyo bacterin prior to challenge with two contemporary virulent field isolates of M hyo. Pigs vaccinated with a M hyo bacterin develop increased serum antibodies, as detected by Tween 20 and DAKO ELISAs, and increased local M hyo-specific IgG and IgA antibodies in BALF at necropsy, compared to nonvaccinated pigs. Vaccination is an effective tool to reduce macroscopic lesions induced by M hyo, although more studies need to be performed using a wider range of field isolates. Acknowledgements The authors thank Nancy Upchurch and students in the Thacker laboratory for their technical assistance. Funding support for this study was provided by Pfizer Animal Health. References 1. Thacker EL, Thacker BJ, Boettcher TB, Jayappa, H. Comparison of antibody production, lymphocyte stimulation, and protection induced by four commercial Mycoplasma hyopneumoniae bacterins. J Swine Health Prod. 1998;6:107-112.

2. Thacker EL, Thacker BJ, Kuhn M, Hawkins PA, Waters WR. Evaluation of local and systemic immune responses induced by intramuscular injection of a Mycoplasma hyopneumoniae bacterin to pigs. Am J Vet Res. 2000;61:1384-1389.

3. Switzer WP. Response of swine to Mycoplasma hyopneumoniae infection. J Infect Dis. 1973;127:Suppl:S59-60.

4. Meyns T, Maes D, Calus D, Ribbens S, Dewulf J, Chiers K, de Kruif A, Cox E, Decostere A, Haesebrouck F. Interactions of highly and low virulent Mycoplasma hyopneumoniae isolates with the respiratory tract of pigs. Vet Microbiol. 2007;120:87-95.

5. Vicca J, Stakenborg T, Maes D, et al. Evaluation of virulence of Mycoplasma hyopneumoniae field isolates. Vet Microbiol. 2003;97:177-190.

6. Strait E, Erickson B, Thacker E. Analysis of Mycoplasma hyopneumoniae field isolates. Proc AASV. Des Moines, IA. 2004:95-96.

7. Artiushin S, Minion FC. Arbitrarily primed PCR analysis of Mycoplasma hyopneumoniae field isolates demonstrates genetic heterogeneity. Int J Syst Bacteriol. 1996;46:324-328.

8. Assuncao P, De la Fe C, Ramirez AS, Gonzalez Llamazares O, Poveda JB. Protein and antigenic variability among Mycoplasma hyopneumoniae strains by SDS-PAGE and immunoblot. Vet Res Commun. 2005;29:563-574.

9. de Castro LA, Pedroso TR, Kuchiishi SS, Ramenzoni M, Kich JD, Zaha A, Vainstein MH, Ferreira HB. Variable number of tandem aminoacid repeats in adhesion-related CDS products in Mycoplasma hyopneumoniae strains. Vet Microbiol. 2006;116:258-269.

10. Stakenborg T, Vicca J, Butaye P, Maes D, Peeters J, de Kruif A, Haesebrouck F. The diversity of Mycoplasma hyopneumoniae within and between herds using pulsed-field gel electrophoresis. Vet Microbiol. 2005;109:29-36.

11. Mayor D, Zeeh F, Frey J, Kuhnert P. Diversity of Mycoplasma hyopneumoniae in pig farms revealed by direct molecular typing of clinical material. Vet Res. 2007;38:391-398.

12. Madsen ML, Oneal MJ, Gardner SW, Strait EL, Nettleton D, Thacker EL, Minion FC. Array-based genomic comparative hybridization analysis of field strains of Mycoplasma hyopneumoniae. J Bacteriol. 2007;189:7977-7982.

13. Bereiter M, Young TF, Joo HS, Ross RF. Evaluation of the ELISA and comparison to the complement fixation test and radial immunodiffusion enzyme assay for detection of antibodies against Mycoplasma hyopneumoniae in swine serum. Vet Microbiol. 1990;25:177- 192.

14. Ross RF, Karmon JA. Heterogeneity among strains of Mycoplasma granularum and identification of Mycoplasma hyosynoviae, sp. n. J Bacteriol. 1970;103:707-713.

15. Friis NF. Some recommendations concerning primary isolation of Mycoplasma suipneumoniae and Mycoplasma flocculare a survey. Nord Vet Med. 1975;27:337-339.

16. Stakenborg T, Vicca J, Butaye P, Imberechts H, Peeters J, De Kruif A, Haesebrouck F, Maes D. A multiplex PCR to identify porcine mycoplasmas present in broth cultures. Vet Res Commun. 2006;30:239-247.

17. Mengeling WL, Lager KM, Vorwald AC. Diagnosis of porcine reproductive and respiratory syndrome. J Vet Diagn Invest. 1995;7:3-16.

18. Thacker EL, Thacker BJ, Young TF, Halbur PG. Effect of vaccination on the potentiation of porcine reproductive and respiratory syndrome virus (PRRSV)-induced pneumonia by Mycoplasma hyopneumoniae. Vaccine. 2000;18:1244-1252.

19. Sambrook J, David WR. Purification of nucleic acids In: Argentine J, ed. Molecular Cloning: a laboratory manual. 3rd ed. Cold Spring Harbor: Cold Spring Harbor Laboratory Press, 2001;A8.9-A8.15.

20. Dubosson CR, Conzelmann C, Miserez R, Boerlin P, Frey J, Zimmermann W, Hani H, Kuhnert P. Development of two real-time PCR assays for the detection of Mycoplasma hyopneumoniae in clinical samples. Vet Microbiol. 2004;102:55-65.

21. Ameri-Mahabadi M, Zhou EM, Hsu WH. Comparison of two swine Mycoplasma hyopneumoniae enzyme-linked immunosorbent assays for detection of antibodies from vaccinated pigs and field serum samples. J Vet Diagn Invest. 2005;17:61-64.

22. Erlandson KR, Evans RB, Thacker BJ, Wegner MW, Thacker EL. Evaluation of three serum antibody enzyme-linked immunosorbent assays for Mycoplasma hyopneumoniae. J Swine Health Prod. 2005;13:198-203.

23. Zielinski GC, Ross RF. Effect of growth in cell cultures and strain on virulence of Mycoplasma hyopneumoniae for swine. Am J Vet Res. 1990;51:344-348.

* Non-referred references

CHAPTER 6. GENERAL CONCLUSIONS General Discussion Mycoplasma hyopneumoniae is one of the most economically important pathogens of pigs and is found throughout the world in most pig-producing countries. It is a primary pathogen and the causative agent of porcine enzootic pneumonia a disease usually characterized by high morbidity and low mortality. Microscopic lesions consist of peribronchiolar and perivascular lymphoid cuffing with infiltrates of inflammatory cells in the airways. Mycoplasma hyopneumoniae is an important component of the porcine respiratory disease complex made up of bacteria such as Pasteurella multocida and Bordetella bronchiseptica and viruses including the porcine reproductive and respiratory virus (PRRSV), swine influenza virus and porcine circovirus 2 (PCV2). Mycoplasma hyopneumoniae not only contributes additively to this complex, but also potentiates the disease produced by both PRRSV and PCV2 by increasing both the severity and duration of disease caused by these viruses. Until recently, it was unknown if all isolates of M. hyopneumoniae were equivalent in their capacity to induce pneumonia. This is due in part to the multi-factorial nature of disease in pigs in the field where a number of organisms and environmental factors affect pigs concurrently. Therefore, assessing the relative contributions of these factors is complicated. In the challenge study comparing four field isolates, differences in the level of virulence among the isolates were observed. Because the virulence factors in M. hyopneumoniae are poorly understood and the protective immune response against this organism has not yet been defined in pigs, we attempted to determine if any marker(s) correlated with increased virulence. While no in vitro or in vivo growth traits or immune parameters measured in this study directly correlated with virulence, two M. hyopneumoniae- specific indirect ELISAs were found to be less sensitive than a competitive-inhibition ELISA at 28 DPC. Much of the genome of M. hyopneumoniae is unique to this organism and identifying conserved genes for PCR assays is difficult. A large number of published PCR assays exist and some of these were developed before the functions of their gene-targets were known. Our findings showed that many of these tests produced false-negative results when DNA 88

from a collection of field isolates was assayed. Because no single currently published real- time assay existed that could detect all isolates of M. hyopneumoniae, two new real-time PCR assays were developed that have detected all isolates tested to date. These assays are a significant improvement over previously described nested-set or real-time PCR assays and provide a means of quantifying M. hyopneumoniae that was not previously available. Three M. hyopneumoniae-specific ELISAs are currently used in the United States. Two are indirect ELISAs manufactured by similar techniques that primarily differ in the strain of M. hyopneumoniae used for antigen production. The third ELISA is a competitive- inhibition ELISA that targets a single internal protein of M. hyopneumoniae. We found the competitive-inhibition ELISA to be more sensitive at earlier time-points than either of the indirect ELISAs in accordance with previous findings. While the two indirect ELISAs performed comparably over all, they differed in approximately one-quarter of the individual results. Further, one field isolate of M. hyopneumoniae was identified that elicited antibodies in pigs that were not detected as well by the indirect ELISAs. These results reinforce the recommendation that M. hyopneumoniae-specific ELISAs should be used on a herd-wide basis and not for individual pig screening. Currently, no M. hyopneumoniae vaccine prevents colonization in pigs from occurring, although vaccines have been shown to decrease the level of pneumonia caused after infection. A challenge study was performed using two recent field isolates of M. hyopneumoniae to determine if variation in isolates affects the ability of a M. hyopneumoniae bacterin to protect against disease. One of these isolates induced approximately twice as many lung lesions as the other isolate. Accordingly, vaccinated pigs that were challenged with the more virulent isolate had increased levels of macroscopic pneumonia compared to vaccinated pigs in the other challenge group. Although for both field isolates, pigs in the vaccinated/challenged groups had significantly less pneumonia than did pigs in the corresponding unvaccinated/challenged groups demonstrating a measure of protection against newer field isolates. Together these studies provide novel in vivo and in vitro comparisons of field isolates of M. hyopneumoniae. They demonstrated variability at the genetic and proteomic levels that impact virulence, diagnosis and prevention of disease in the pig. 89

Recommendations for Future Research No traits were definitively correlated to specific virulence factors in high versus low pathogenic isolates. Because pig-to-pig variation is common in M. hyopneumoniae challenge studies, future studies should include increased numbers of pigs. The isolates used in all of our studies came from clinical cases submitted for necropsy. It may be beneficial to identify more isolates from herds that are subclinical in order to more easily identify distinct traits associated with virulence. Other technologies targeting genetic (microarray) and proteomic (MALDI-TOF) differences should also be employed. Diagnostic assays for M. hyopneumoniae are hampered by the unique genetic and immune characteristics associated with the organism and the disease it causes. As more information concerning gene function and virulence factors becomes known, improvements in the different assays can be made. Two new real-time assays were developed and have detected all isolates of M. hyopneumoniae tested in our collection. These assays should be further validated using isolates world-wide. This study is already underway. Future studies should also work to improve ELISAs for M. hyopneumoniae. A new ELISA should target more than one protein to guard against false-negatives and the antigen should be well-defined rather than a crude protein preparation to guard against false-positive results. A study with these aims has already been outlined in a grant submission. All vaccines for M. hyopneumoniae are bacterins. No modified-live or sub-unit vaccines exist. Future studies may be able to identify candidate proteins for sub-unit, or DNA vaccines. In the short-term, it may be useful to update the strains used in the killed vaccines, or include a mixture of strains in order to provide improved protection. Challenge studies should be performed using field isolates with a wide range of virulence and genetic diversity in order to assess protection. Future studies should more clearly define the immune response in pigs infected with M. hyopneumoniae. A microarray of immune-related genes in the pig is being employed to do this. Techniques such as tracheal epithelial explants, and laser-capture microdissection will aid in identifying the specific contributions of individual cell types to the disease process as well as to the immune response.

ACKNOWLEDGEMENTS I would like to acknowledge the many people who have contributed to my research and professional development during my time as a graduate student. I thank Barbara Erickson, Nancy Upchurch and Theresa Young for their willingness to share their vast technical knowledge and skills. They have each been an invaluable resource and I greatly appreciate their work on my behalf. Fellow graduate students in the laboratory: Pravina Kitikoon, Amy Vincent, Shan Yu, Katherine Jones and Dachrit Nilubol, deserve many thanks for their friendship and helpful insights that made the lab and enjoyable and productive place to be. I also want to thank the many students that have worked in the laboratory over the years. They are too numerous to name, but I especially want to thank Gavin Yager, Margaret Aleff, Cassandra Roe and Katie Behrens for their considerable help in completing my projects. In addition to industry support from several companies who performed studies with our laboratory, grants from the National Pork Board, the Iowa Pork Producers Council, the USDA Formula Fund, the Healthy Livestock Initiative, and the Iowa Livestock Health Advisory Council made my research possible through their financial support. Members of my committee: Drs. Eileen Thacker, F. Chris Minion, Brad Thacker, Patrick Halbur and Douglas Jones have all been generous with their time and support. I especially thank my major professor, Dr. Eileen Thacker. She has been a tremendous mentor to me and provided me with the tools to grow as a scientist. In addition, Dr. F. Chris Minion deserves special recognition for all the time, energy and personal attention he has devoted to me and my professional development. Dr. Melissa Madsen has been an invaluable resource. Her friendship and collaborative efforts are greatly appreciated. Drs. Bruce Janke, Brad Van De Woestyne and Jane Christopher-Hennings deserve recognition for their contributions to my studies. I also thank the Iowa State University Veterinary Diagnostic Laboratory for all the cooperation and expertise they have provided over the years. Finally, I acknowledge the support given to me by my family. I thank my dad for his never-ending encouragement. My sister, Ashley, has been a wonderful friend and a great source of moral support. Brad, my husband, has been more than supportive during my years as a “professional student”. I thank him for his love and sacrifice as I pursued my academic goals. Lastly, I am grateful to my children Colin and Jillian who brighten my days and provide my life with perspective.