The past, present and future of the control of M

The past, present and future of the control of M. hyopneumoniae infection Chris Morrow PhD, BVSc Honorary Associate Professor, The University of Melbourne Technical Manager, Bioproperties, Australia [email protected]

Resumen Un estado de enfermedad mínimo se ha sugerido como una estrategia para limitar el impacto de las infecciones en la industria porcina por exclusión. Esto implica que al menos que la infección sea autolimitante o pueda ser eliminada por una intervención, los animales infectados (las piaras completas) deban ser sacrificadas. Las estrategias para lograr el concepto de enfermedad mínima, tienen diferentes grados de adopción, dependiendo en parte de los costos de una bioseguridad efectiva, el costo de la despoblación, el riesgo de la infección (o la reinfección), la habilidad para obtener una fuente animal no infectada y el costo de vivir con la infección y con la enfermedad asociada. Estos costos varían con el tiempo y la eficacia de formas alternativas de cómo hacer frente a la infección también puede cambiar. No todas las partes integrantes de la Industria Porcina han adoptado las estrategias de enfermedad mínima por las razones mencionadas anteriormente. Esto ha conducido a estrategias modificadas como la “despoblación Suiza” y la “Toma al nacimiento” con un mayor riesgo pero un menor costo. Utilizar a los antibióticos como profilaxis es una estrategia específica contra bacterias (no considerado como un promotor de crecimiento). En general, muchos antibióticos pueden ser efectivos contra un gran espectro de problemas de origen bacteriano (en orden de importancia decreciente en la industria porcina mundial: Mycoplasma hyopneumoniae y otros, Lawsonia, Pleuroneumonía porcina, Espiroquetosis intestinal porcina, Streptococcus suis, Haemophilus, Pasteurella, Salmonella, Erysipelothrix, etc.) a menudo ha ayudado a la industria a manejar estas infecciones. Por una variedad de razones técnicas, incluyendo la resistencia antibiótica y la presencia de los residuos antibióticos, hay ahora presiones para reducir la dependencia a los antibióticos. También el costo financiero de la profilaxis con antibióticos debe de ser considerado. La vacunación como intervención ha sido sumamente efectiva en limitar los efectos de los desafíos virales. Las bacterinas (incluyendo a las autógenas) han sido utilizadas por un largo tiempo en la industria porcina contra varios patógenos y son muy buenas en detener los efectos de las enfermedades bacterianas, produciendo una inmunidad de tipo humoral (anticuerpos). La protección proporcionada por la inmunidad humoral en ocasiones es limitada al desafío por organismos de la misma serovariedad (homólogos). Las vacunas vivas bacterianas no han tenido un mayor impacto en la industria porcina. Sin embargo, estas vacunas pueden generar otros tipos de inmunidad así como la inmunidad humoral y ofrecer una protección de más amplio espectro, incluso entre especies. Este trabajo describirá a la nueva vacuna viva de Mycoplasma hyopneumoniae y tratara de predecir cómo podrá ser utilizada en la industria porcina moderna, basándonos en las experiencias similares que hemos tenido con vacunas vivas contra micoplasmas aviares en la industria avícola.

Summary Minimal disease status has been suggested as a strategy to limit the impact of infections in the swine industry by exclusion. This implies that, unless infection by the pathogen is self-limiting or can be sterilized by intervention, infected animals (herds) have to be slaughtered. Minimal disease strategies have had variable degrees of adoption, depending in part on the costs of effective biosecurity, the cost of depopulation, the risk of infection (or re-infection), the ability to source uninfected stock and the cost of living with the infection and associated disease. These costs vary

over time and the effectiveness of alternative ways of coping with the infection also can change. Not all parts of the swine industry have adopted minimal disease strategies for the reasons discussed above. This has also led to modified strategies like the “Swiss depop” and “snatch farrowing”- higher risk but lower cost. Antibiotic prophylaxis is a specific intervention against bacteria (not a feed ingredient that improves performance for a while). In general, many antibiotics can be effective against a broad range of bacterial problems (in order of decreasing importance in the swine industry world-wide:- Mycoplasma hyopneumoniae and others, Lawsonia, porcine pleuropneumonia, Porcine Intestinal Spirochaetosis, Strep. suis, Haemophilus, Pasteurella, Salmonella, Erysipelothrix etc) and has often helped the industry to manage with these infections. For a variety of technical reasons, including acquired resistance and residues there are now pressures to decrease dependence on antibiotics. There is also the financial cost of antibiotic based prophylaxis to be considered. Vaccination as an intervention has been very successful in limiting the effects of viral challenges. Killed bacterins (including autogenous bacterins) have been used for a long time in the swine industry against various pathogens and are very good at stopping effects of bacterial diseases usually by inducing humoral (serum) antibody. The protection provide by humoral antibody is often limited to challenge by the organisms of the homologous serovar. Live bacterial vaccines have not had a major impact in the swine industries. However these vaccines can generate other forms of immunity as well as humoral immunity and often offer broader protection (broader than serovar classification and even between species). This paper will describe a new live Mycoplasma hyopneumoniae vaccine and predict how it might be used drawing in the modern swine industry based on experiences with similar live mycoplasma vaccines in poultry industries. Introduction Mycoplasma hyopneumoniae (Mhyo) is a respiratory infection of pigs that under certain circumstances may cause respiratory disease and poor performance in infected animals especially in the grower-finisher period. This organism chronically infects the pigs for life but naturally developing immunity may decrease the number of organisms over time and this may have effects on transmission as well as clinical effects and performance. Stress or other respiratory infections may cause a recrudescence and increase in numbers being excreted. There may be large variation in pathogenicity between strains especially where partial control programmes have been used. Electron microscopy shows large numbers of Mhyo organisms adherent to the respiratory tract cilia, which can lead to deciliation and impairment of muco-ciliary. Natural transmission on infected farms is thought to happen when piglets are sharing the same airspace as infected sows but also can occur between weaners in the grower phase. Biosecurity is important for the control of Mhyo infection. Airborne spread over distances five kilometres has been demonstrated but the risk is diluted as distance increases. Indeed the Achilles heel of Mhypo minimal disease strategies is that animals free from infection have no resistance to becoming infected when challenged. Our current farm setup is often suboptimal with farm units being too close together (veterinary advice to move an established farm is usually seen as academic and impractical). Introduction of infected stock is a common cause of biosecurity breakdown. The organism can also be moved by people and on fomites for short periods of time. If we could make pig herds, more resistance to these field challenges this would be a big advance. Antibiotics against mycoplasmas (most except those active against bacterial cell wall synthesis; penicillins, cephalosporins, and phosphomycin) are commonly given in pig feed to enable farms to live with the infection. It is generally agreed that antibiotics will not cure the infection (or all mycoplasma problems in all animal industries would have been solved in the 1950s). Acquired antibiotic resistance is an additional problem that is reasonably believed to be directly proportional to the intensity of antibiotic use, and cross resistance within antibiotic groups is significant. Antibiotic sensitivity testing of Mhyo isolates is doubly difficult – isolation being difficult and subsequent Minimal Inhibitor Concentration testing also being technically demanding. Prophylactic antibiotics can also contribute to control of problems caused by other porcine mycoplasmas, Lawsonia, Porcine pleuropneumonia, Porcine Intestinal Spirochaetosis, Strep. suis, Haemophilus, Pasteurella, Salmonella, and Erysipelothrix if these are present on the farm or challenging from farm surroundings. Similarly vaccines (including autogenous vaccines) are used to help control these problems. Indeed removing antibiotics may see the emergence of diseases from infections that were not even suspected to be on the farm.

Killed Mhyo vaccines are also used to decrease the effects of Mhyo infection (lung pathology and production efficiency). Sometimes in combination with antibiotics (often it is hard to tell which intervention is producing the control). These vaccines produce humoral immunity to Mhyo through the production of serum antibodies. They do not stop the infection of pigs and do not appear to alter the dynamics of Mhyo infection. Similar remarks probably apply to bacterins (including autogenous ones) for other swine pathogens. Duration of immunity from killed vaccines, unless boosted, may not be sufficient to be useful. Killed vaccines are delivered by needle which can break off in the pig causing quality problem. A live Mhyo vaccine may have certain advantages over killed vaccines and antibiotics. This paper describes where such a vaccine may provide benefits in farms where Mhyo cannot be reliably excluded. These predictions are based on experience with similar vaccines in the poultry industries. Experience with poultry attenuated poultry mycoplasma vaccines. The ts-11 strain Mycoplasma gallisepticum (MG) vaccine and the MSH strain M. synoviae (MS) vaccine were developed in the 1980s at the University of Melbourne under the supervision of Dr Kevin Whithear. This was in response to a problem in the poultry industry where tylosin had stopped working against MG in broiler production. The investigations to control the MG problem had progressed along a linear development pathway from first the demonstration that field strains of MG had acquired tylosin resistance, then evaluating killed MG vaccines by injection, killed antigen given intranasally, using a mild strain of MG as a vaccine (similar pathogenicity to F strain), and finally, inspired by a Japanese idea, the attenuation of mycoplasma strains by exposure to a mutagen and selection of clones for temperature sensitivity (a greater than 100 fold reduction in titre when grown at the non-permissive temperature) resulted in development of an attenuated vaccine candidate. This vaccine candidate was not a GMO as it was developed using classical bacteriological techniques. Ts-11 has gone on to become one of the most successful mycoplasma vaccines in poultry and the first and only one to be used in Breeders. Later the MS vaccine MSH was developed by me using a similar pathway. Temperature sensitive mutants were expected to only be able to colonize the upper respiratory tract, but to be able to generate mucosal immunity which would disseminate to other mucosal surfaces. Indeed this seems to be the case. Sometimes on re-isolation the vaccine progeny had lost its temperature sensitive phenotype but it had not re-acquired pathogenicity when assessed experimental infection models. We now understand that typically 30 to 100 genomic changes are present in those temperature sensitive mutants that can colonise animals compared to their parent strains. Indeed the temperature sensitivity of the attenuated MS vaccine MSH is thought to be a single base change in the obg gene but this is not the case in all these temperature sensitive mycoplasma vaccines. The acquisition of temperature sensitivity by clones after mutation of some surviving organisms demonstrates that those clones have been modified by the mutagen and implies that other genotype changes have been also induced in those clones. This has been confirmed by sequencing studies. The responses to these vaccines by the animal are slower than the responses to infection with the parent strains, where this has been characterized. The decreased pathogenicity, the slower onset of immunity, the slower seroconversion and the reduction in horizontal transmissibility are examples of phenotypic differences. So perhaps it is the partial crippling of the vaccine strain by mutagenesis that allows the vaccinated animal to always subdue the vaccine; a tipping of the balance in favour of the vaccine. Initially this may be the temperature sensitivity, but other (yet uncharacterized) genomic changes strengthen the attenuation. Like live viral vaccines it is important to vaccinate the whole flock with a full dose of vaccine so animal-to-animal transmission of the vaccine is prevented. Later the development of immunity will control this transmission. Pseudo-revertants of ts-11 and MSH that have lost their temperature sensitivity have not reacquired their pathogenicity. In the last twenty years these vaccines have worked in all places they have been tried around the world against field challenges by local strains. Mycoplasma species are defined by serological relatedness to type strains, similar to Newcastle disease virus, so there is only one serotype by definition. No examples of vaccination failure by challenge by “variants” have been described. We have evidence that protection induced by these vaccines is not humoral antibody and indeed in some vaccinated flocks systemic antibody against the organism may be very difficult to detect. Birds from such flocks have been taken back to the laboratory and challenged and shown to be protected from mycoplasma disease. Attempts have been made

to improve the sensitivity and quality of serological tests by using cloned homologous antigens from the vaccines. Improvements were seen but still a large variation in responses in flocks (and birds) make this hard to interpret. Humoral antibody seems to be generated by the vaccines more in birds with tracheitis around the time of vaccination and titres may be more a measure of tracheitis than vaccine response and protection. Certainly we have seen in the same experiments that killed vaccines producing large humoral immune responses did not protected against challenge. It is interesting to note that broiler breeders seem to produce less of an antibody response to the vaccines than layer breeds. The serological responses of flocks from six weeks after vaccination with ts live mycoplasma vaccines fall into four basic patterns 1) Moderate levels of antibody – originally thought to be the only expected response 2) Low to no detectable antibody (these are the ones we investigated by bringing birds back to the laboratory and challenging and demonstrated that they are indeed immune) 3) Initially low levels of antibody, later (usually from the beginning of lay to 40 weeks) rising to high levels. Investigation in these flocks with strain identification PCR surveys has not identified field infection. 4) Different patterns from the previous flocks – this is often seen after use of the vaccines for a number of years where flocks suddenly remain seronegative after vaccination. Perhaps this is when wild strain challenge has been completely displaced. These responses are difficult to interpret – in contrast, field challenge usually causes rapid development of high antibody levels, although even this is not universal with all strains, and the antibody response to field challenge may also be decreased by effective antibiotic administration. Monitoring of vaccination is by auditing the administration of the vaccine and by training of staff (as antibody production is so variable). The addition of dye to the vaccine can confirm eyedrop vaccination- evaluation is by assessing tongue staining in the next 30 minutes. Investigation of suspect problems (clinical signs, poor hatchability and other production parameters) in vaccinated flocks is by PCR and in some cases culture. DIVA (Differentiation of infected from vaccinated animals) is possible with DNA based assays. Strain identification can be done straight from swabs without sacrificing the animal. The best targets are those that are mycoplasma species specific (surface proteins – adhesion molecules) but span an area where there is a lot of intra-species variation. Thus a positive PCR confirms specific mycoplasma infection, while differentiation between vaccine and field strains is assessed by further comparison of the PCR product’s sequence. This can be done by sequencing, internal labelled probes or HRM (high resolution melting analysis). Thus results may be obtained very rapidly. In the latter two cases the techniques can be done directly from swabs and completed immediately. All these PCR tests have various advantages and disadvantages – especially in terms of speed of result, utility of the information generated, and the ability to identify more than one strain in an assay. Horizontal transmission of ts-11 and MSH appears very limited. “Inadvertent vaccination” has been described, with the transmission of the vaccines to unvaccinated flocks, and the explanation has been that vaccinating teams have transferred the vaccine to the non-vaccinated farm. ts-11 has not been found to persist on multi-age poultry farms (in contrast to mild strains). Horizontal spread of vaccine strains does occur when vaccinated birds are mixed with uninfected unvaccinated birds (for example “spiking” flocks with vaccinated males). Horizontal transmission occurs readily with beak to beak contact but often transmission does not occur from row to row in cage houses and very rarely from shed to shed. Farm to farm transmission by airborne routes has not been observed. Certainly for the poultry industries this level of protection is very useful and has allowed vaccinated flocks to remain free from infection with field strains and the eradication of field strains from multi-age farms has been achieved with only a minimum level of biosecurity. Vertical transmission of MG and MS inside the egg has been stopped by vaccination but this is aided by the precocial nature of the chicken which allows regular complete separation of generations. Mhyo transmission is thought to occur from the sow as the piglets suckle, but also horizontal transmission (challenge) in grower sheds is also very important. No benefit can be expected by vaccination of birds that are already infected but vaccination of young stock (at 3 weeks of age) derived from infected parent stock has been quite successful where the birds have been “flushed” with antibiotics a week before vaccination where a large proportion of the flock has not yet become infected.

Routine antibiotic prophylaxis programmes that have been employed by the poultry industries are one of three main programmes 1) Treatments in the first week of life – non-specific, probably providing some control against a broad range of bacteria, including mycoplasmas. 2) Regular treatments of layers or breeders every four to eight weeks (this is targeted at mycoplasma infections). 3) Regular treatments of (vertically infected) broilers at 20-22 days of age (or sometimes earlier starting at day18 to control post vaccinal reactions to strong Newcastle disease vaccines). This is to prevent chronic respiratory disease from day 26 onwards. Although there is no useful transmission of vaccinal immunity to progeny in chickens (maternal antibody actually may increase vertical transmission by allowing more infected chicks to hatch) the advantages of having mycoplasma negative day old chicks from vaccinated flocks can be exploited to reduce dependence on antibiotics in the broiler generation. The mechanism of action of these vaccines in chickens is the generation of mucosal immunity. We have not been able to pin down the exact mechanism of immunity but it is not humoral antibody or competitive exclusion. Mucosal immunity is usually short lived and for maintenance needs constant antigenic stimulation with small amounts of antigen (for example successful coccidiosis vaccination). The survival of the vaccine strain in the upper respiratory tract provides this stimulation. In Iran 93% of vaccinated flocks can still be shown to contain the vaccine strain at 68 weeks of age. The relationship between mycoplasmas and the host is very complex, with the mycoplasmas possessing many complex mechanisms for maintaining chronic infection. The kinetics of the development of the vaccine population in a bird and a flock may influence the onset of immunity and stability of the vaccine in the field. For this reason a full dose needs to be given to all birds at the time of vaccination. MSH has been used to stop Infectious synovitis in South Africa and decrease antibiotic use totally in rearing of pullets in some operations. It has also been used against Egg Apical Abnormality in Japan and Europe again decreasing antibiotic dependence. Ts-11 in Australia and many other places in the world has eliminated respiratory problems and decreased antibiotic usage in the vaccinated generation and often in their progeny – most effectively most often in combination with MSH vaccine. Production responses have often been greater than previous estimates for the cost of infection. For example MSH vaccinated flocks compared to unvaccinated but infected flocks have about 4% better feed conversion in the production of eggs. Other observations have included decreased peritonitis in vaccinated breeders. Clinicians report that in vaccinated flocks where the effects of viral and bacterial challenges (non-mycoplasmal) are ameliorated, becoming uncomplicated infections resolving faster rather that becoming chronic respiratory diseases; again decreasing dependence on antibiotics. The success with these vaccines in poultry production on national industry scales gives us a model for how such a similar vaccine may be used in the swine industry. Vaxsafe MHP (strain ts-19). Vaxsafe MHP was created by chemical mutagenesis of the M. hyopneumoniae parent strain LKR and selection of mutants that had decreased ability to grow at 39.5°C compared to 33°C. In the case of ts-19 this reduction in titre is absolute. Worldwide patents on the vaccine strain have been applied for, along with further patents on the genomic changes in the vaccine from the mutagenesis. The strain was seed-lotted and vaccine manufacture optimized, with the final product being freezed dried and transported on dry ice. The vaccine was shown to be safe in standard tests at X1 and X10 doses. It passed the international standard reversion to virulence test (VICH 41). Duration of immunity is yet to be established but is expected to be for life while the animal is still infected with the vaccine strain. For this reason we also think that revaccination would not be necessary (although it may stimulate more antibody- perhaps of comfort to veterinarians unconvinced by the previous arguments). Piglets are vaccinated by the intra-nasal route. The vaccine was shown to be effective in protecting against in-contact challenge using an Australian field isolate when given to piglets in the first week or at three weeks of age (Challenge being 5 weeks after vaccination). Protection against deciliation in the upper respiratory tract was seen using scanning

EM to examine samples taken from vaccinated and challenged pigs in Australian pen and field trials. Field trials in Mexico showed the vaccine to be efficacious when piglets were vaccinated in the first week of life. Vaccination resulted in decreased lung lesion scores and increased daily weight gain. The strength of the challenge in this last field trial was moderate, with the solid pen walls not allowing “snout-to-snout” contact of infected pigs with vaccinated pigs. Japanese studies have demonstrated that the efficacy of ts-19 vaccination against a local isolate challenge (pneumonic lesions and production effects) was not affected by maternal antibody in piglets vaccinated at one day of age. No Mhyo ELISA antibodies could be detected in the colostrum deprived piglets at 3 weeks of age, in contrast to the colostrum fed group but the performance and lung lesions after challenge was the same for both groups at 56 days of age. This suggests maternal antibody (mostly IgG) does not interfere with using ts-19 vaccination. The usefulness of serology in monitoring swine herds vaccinated with ts-19 needs further investigation but based on experience in chickens protection will probably not be correlated with serum antibody levels. Currently two ELISAs (Biochek and ID-Vet1) use cloned Mhyo antigen from the LKR parent of strain ts-19 in their kits and these should have better performance when monitoring vaccinated herds for antibody, but whether any serologically based DIVA strategy can be developed is yet to be seen. With MSH vaccine, some poultry veterinarians have used two ELISAs – one with a cloned antigen from the vaccine and another with whole cells, and shown that the usual response of vaccinated flocks is to seroconvert at 4 weeks post-vaccination as detected with the cloned MSH antigen while seroconversion with the whole cell ELISA is delayed. The onset of immunity needs further characterization, but the trials to date have challenged piglets as early as 4 weeks after vaccination and protection was demonstrated. Whether maximal immunity has developed by this time is not yet known. If possible vaccination should be undertaken in an antibiotic free window and antibiotic use minimized for the next four weeks after vaccination (or preferentially select antibiotics with no anti-mycoplasmal activity or use in feed antibiotics that are not absorbed from the gut). In the live animal samples from nasal swabs of pigs can be used to assess infection status. A DIVA test can be developed from the genome differences that have been identified from the LKR parent strain and ts-19.

Advantages for the swine industry The ts-19 vaccine is not just a super killed vaccine that is given without a needle. It is a novel approach to controlling Mhyo infections. Killed Mhyo vaccines have allowed us to live with Mhyo infection. Ts-19 has the potential to let us live without M. hyopneumoniae field strain infection, even in situations where challenge is always occurring. Porcine respiratory disease appears to be very similar to Chronic Respiratory Disease in the chicken. In both cases the mycoplasma components are crucially important as they aggravate respiratory damage and allow secondary pathogens to cause havoc. The control of Mhyo infection by vaccination will be the first step in reducing antibiotic dependence of the swine industry for those farms where biosecurity cannot be further upgraded. In the poultry industry these vaccines have been able to protect birds from becoming infected with field strains of MG and MS. This in turn has led to massive reductions in the need to use antibiotics in these production systems where challenge is limited to airborne (dilute) exposure. Induction of humoral antibody is not necessary for effective protection against challenge. Colonization of the upper respiratory tract appears to be the important factor for induction and maintenance of immunity. There is some evidence that the immunity induced by killed vaccines may interfere with the generation of maximal immunity from the live vaccines. Certainly, poultry sites that have had the best results are those that do not also use killed vaccines and have removed routine antibiotic prophylaxis programmes (although they still treat when necessary). All of these vaccines are sensitive to all anti-mycoplasmal antibiotics whereas field strains may have acquired resistance.

1

ID Vet www.id-vet.com/produit/id-screen-mycoplasma-hyopneumoniae-indirect Biochek www.biochek.com/mhyo-mycoplasma-hyopneumoniae-antibody-test-kit.html

In the swine industry this vaccine will be able to be integrated into a total strategy for control of the impact of Mhyo and reducing antibiotic dependence . This may be 1) On infected farms- preferably where mixing of vaccinated and unvaccinated pigs is minimized 2) On farms being repopulated; to protect the investment in establishing Mhyo (and perhaps Actinobacillosis)free herds using “low tech” snatch farrowing 3) Establishing sow operations where eventually all sows are vaccinated with ts-19. In this situation we can expect to see a cumulative improvement of their Mhyo status and health over time as the total displacement of field strains progresses. Of course the power of the vaccine to deliver benefits will be improved on farms where batch grow out is practised. The sourcing of Mhyo free sows will also be important and these can be protected before challenge by vaccination. Strategies for control of other bacterial problems to decrease antibiotic dependence and the minimization of the effects of viral infections also need to be considered. We see vaccination not as part of a strategy for eradication and moving from infected to uninfected status and stopping vaccination but as a management tool to continuously protect the uninfected status of a herd from the challenges arriving from neighbouring farms and other breaks in biosecurity. Acknowledgments This is the culmination of over a decade of research by a cast of thousands. LKR was characterized by Len Lloyd and Geoff Cottew (CSIRO). Kevin Whithear and Phil Markham led the production of ts-19 vaccine candidate (The University of Melbourne). Technically they were helped by Anna Kanci and others. The development of this vaccine candidate into a vaccine was by Youssef Abs El-Osta, Rima Youil and team at RMIT, manufacture was scaled up by Julie Brogestam and others at the Glenorie Manufacturing facility. Horacio Lara, CENASA, and Yukio Seiyka for field trails in Mexico and Japan. Greg Underwood, Glenn Browning and Ross Henderson have made special contributions. David and Anthony of Bioproperties for the money (and other funding bodies). If I have seen further than others it has been from standing on the shoulder of giants.