Clinical expression, epidemiology, and monitoring of Mycoplasma gallisepticum and Mycoplasma synoviae: an update

ABSTRACT

Mycoplasma gallisepticum (MG) and Mycoplasma synoviae (MS) are of clinical and economic importance for the global poultry industry. Many countries and integrations are involved in monitoring programmes to control both mycoplasma species. This review provides an extensive historic overview of the last seven decades on the development of the knowledge regarding the factors that influence the clinical expression of the disease, the epidemiology, and monitoring of both MG and MS. This includes the detection of new virulent strains, studies unravelling the transmission routes, survival characteristics, and the role of other avian hosts. Also the role of molecular typing tests in unravelling epidemiology and factors that complicate the interpretation of test results is discussed. The latter includes the presence of heterologous mycoplasma infections, the use of heterologous oil-emulsion vaccines, and the use of antibiotic treatments. Also the occurrence of MG and MS strains with low virulence and the use of live and/or inactivated MS and MS vaccines are discussed.

KEYWORDS:

• Mycoplasma gallisepticum
• Mycoplasma synoviae
• expression
• epidemiology
• monitoring
• review

Introduction

The clinical and economic importance of Mycoplasma gallisepticum (MG) in the poultry industry was already acknowledged in the early 1950s (Floyd et al., 1952; Crawley & Fahey, 1955). For Mycoplasma synoviae (MS), the clinical impact for the poultry industry was also recognized in early reports (Olson, Adler et al., 1964). Since then, an increase of economic impact has been recognized due to the identification and emergence of new pathogenic strains (Kerr & Olson, 1969; Morrow, Bell et al., 1990; Landman & Feberwee, 2001, 2012; Landman & Bronneberg, 2003). The first description of avian mycoplasmas dates from 1936 (Nelson, 1936) and was related to chickens with respiratory disease from which coccobaciliform bodies were isolated (Nelson, 1935); they were later renamed pleuropneumonia-like organisms (Barber & Fabricant, 1962). This respiratory disease was later defined as chronic respiratory disease (CRD) in chickens and infectious sinusitis in turkeys (Floyd et al., 1952), and the aetiological agent identified as MG (Edward & Kanarek, 1960). Infectious synovitis was reported for the first time in the 1950s (Olson et al., 1954; Chalquest & Fabricant, 1960) and the aetiological agent was later identified as MS (Olson, Adler et al., 1964; Olsen, Kerr et al., 1964). Most important was the evidence that both mycoplasma species could exhibit a variance in clinical expression (Ghazikhanian et al., 1973; Lin & Kleven, 1982a), especially in the presence of other respiratory agents, and that synergistic effects with other respiratory agents had negative effects on poultry performance (Fahey & Crawley, 1954b; Gross, 1961; Corstvet & Sadler, 1966; Springer et al., 1974; Rhoades, 1977). Studies unravelling their transmission routes and epidemiology contributed to the development and implementation of the first monitoring and control programmes for MG (Crawley & Fahey, 1955; Chute et al., 1965). The knowledge regarding the expression of clinical disease, epidemiology, and the implementation of monitoring programmes was crucial for the control of both species in the commercial poultry industry in the last six decades, and contributed to the economic sustainability of the poultry industry of today. This paper gives an overview on the historical and current knowledge regarding clinical expression of the disease, the epidemiology, and monitoring of MG and MS.

Clinical expression

Clinical expression of MG

MG is associated with respiratory disease, egg production losses, reduced hatchability, and significant downgrading of carcasses at slaughter due to the systemic effects (Stipkovits & Kempf, 1996; Kleven, 1997; Levisohn & Kleven, 2000). Besides commercial poultry, clinical outbreaks in backyard poultry and other avian species including wild birds have also been reported (Madden et al., 1967; Cookson & Shivaprasad, 1994; Ley et al., 1996; Benčina et al., 2003). In the early 1950s it was shown that MG was responsible for turkey sinusitis, CRD in chickens, egg production losses up to 50% during 7–14 days not recovering up to the expected production level, reduced eggshell quality, and reduced hatchability due to late embryo mortality in chickens (Markham & Wong, 1952; Fahey & Crawley, 1954b). Other observations confirmed that MG alone was able to induce disease and that the expression of clinical signs caused by MG varied greatly between infections with different isolates (Gross, 1961; Truscott et al., 1974; Jordan, 1975; Rodriguez & Kleven, 1980; Yoder, 1986). Also, strains with differences in tissue tropism and/or virulence are described such as the MG R strain, MG S6 strain, and WVU 907 strain. The MG R strain is a strain isolated from turkeys with sinusitis (Markham & Wong, 1952), while the MG S6 strain was isolated from the brains of turkeys with infectious sinusitis and neurological signs (Zander, 1961). More evidence of the involvement of MG in neurological signs in turkeys was reported over the years (Thomas et al., 1966; Chin et al., 1991; Wyrzykowski et al., 2013). Furthermore, MG is also reported in relation to joint disease (Wannop et al., 1971; Morrow et al., 1997). In 1974 for the first time, a MG strain of low virulence was reported (Truscott et al., 1974). This strain induced a low number of reactors in serological tests, while no clinical signs were observed (Truscott et al., 1974; Yoder, 1986; Dingfelder et al., 1991). In 1981, the Mycoplasma strain WVU 907 was designated as a MG strain of low virulence (Sahu & Olson, 1981; Yoder, 1986; Kempf et al., 1997).

Clinical expression of MS

The clinical and economic relevance of MS for the commercial poultry industry was evidenced in several early studies. MS monitoring and control programmes have already been running in reproduction stock for a long time and include culling of MS-positive flocks and/or streaming of infected products. However, in the last two decades, the identification of new virulent MS strains, with economic impact on poultry production, increased the attention for the control of MS in the commercial poultry industry in general (Landman, 2014). Different from MG, until now only subclinical infections have been reported in backyard poultry and other wild free-flying birds such as pigeons, crows, and sparrows (Benčina et al., 1988a, 1988b; Bradbury et al., 2001b; Michiels et al., 2016). MS is known as the cause of infectious synovitis, respiratory disease, egg production losses, and eggshell abnormalities in commercial poultry. The first report involving MS was related to infectious synovitis (Olson, Kerr et al., 1964). Since this finding, MS infection has been widely reported as the causative agent of infectious synovitis in turkeys and chickens (Kerr & Olson, 1969; Morrow, Bell et al., 1990; Landman & Feberwee, 2001, 2012; Landman & Bronneberg, 2003). Also, the involvement of MS in respiratory disease has been shown in several studies. MS was recovered from the respiratory tract of birds without clinical disease (Olson, Adler et al., 1964), while other studies showed that MS infections were related to egg production losses, airsacculitis in broilers and turkeys, and reduced broiler performance (King et al., 1973; Vardaman, Reece et al., 1973; Bradbury & Howell, 1975; Goren, 1978; Lott et al., 1978; Macowan et al., 1982; Osorio et al., 2007). Moreover, tissue tropism seemed relevant as studies showed that some MS isolates were more related to the presence of joint lesions, while other isolates were more related to the presence of lesions in the respiratory tract (Kleven et al., 1975). In 2009, a MS isolate related to egg production losses, eggshell apex abnormalities (EAA), and reduced eggshell strength was reported for the first time (Feberwee et al., 2009). Broiler-breeder birds seemed less susceptible for the production of eggs with EAA than commercial layers (Feberwee & Landman, 2010). From 2009 onwards, several reports on MS related to EAA, egg production losses and/or reduced eggshell quality, and increased second-quality eggs followed and showed the worldwide presence of such strains (Catania et al., 2010; Ranck et al., 2010; Gole et al., 2012; Jeon et al., 2014; Kursa et al., 2019; Lorenc et al., 2019; Cisneros-Tamayo et al., 2020). Moreover, other reports also evidenced the amyloid-inducing potential of MS (Landman & Feberwee, 2001).

Factors influencing clinical expression MG and MS

The role of other respiratory pathogens, including viral vaccine strains, in aggravating disease or the expression of CRD in the presence of inapparent MS and MG infections, was acknowledged as early as 1955 (Markham & Wong, 1952; Crawley & Fahey, 1955; Gross, 1961). During the last four decades the significant role of several respiratory agents complicating disease induced by MG has been confirmed. The synergistic effect of MG with other bacteria, such as Haemophilus gallinarum (Avibacterium paragallinarum), Mycoplasma meleagridis, Pasteurella gallinarum, and Escherichia coli, was evidenced (Gross, 1958; Kato, 1965; Reis & Yamamoto, 1971; Chu & Uppal, 1975; Bradbury, 1984; Droual et al., 1992). Other studies evidenced the role of respiratory viruses including vaccine strains such as Newcastle disease virus (NDV), infectious bronchitis virus (IBV), and avian metapneumovirus in the aggravation of the disease in the presence of MG (Corstvet & Sadler, 1966; Sato et al., 1970; Nonomura & Sato, 1971; Omuro et al., 1971; Rodriguez & Kleven, 1980; Naylor et al., 1992; Leigh et al., 2012). The most recent studies show the synergistic effect between a H3N8 low pathogenic avian influenza virus and MG (Stipkovits, Egyed et al., 2012; Stipkovits, Glavits et al., 2012). Respiratory bacteria (such as M. meleagridis, Pasteurella gallinarum, E. coli, O. rhinotracheale), and respiratory viruses, including vaccine strains, have also been shown to complicate disease induced by MS (Kleven et al., 1972; Springer et al., 1974; Bradbury & Howell, 1975; Kleven, 1975; Rhoades, 1977; Goren, 1978; Hopkins & Yoder, 1984; Droual et al., 1992; Khehra et al., 1999; Zorman-Rojs et al., 2000; Landman & Feberwee, 2004; Raviv et al., 2007; Feberwee et al., 2009; Feberwee & Landman, 2010; Roussan et al., 2011). In a recent paper, a frequent association of detection of M. pullorum and MS was found in cases with eggshell apex abnormalities (Cisneros-Tamayo et al., 2020). However, further studies have to confirm the synergistic role of M. pullorum. The degree of synergism between MS and IBV strains seems to be influenced by the virulence of the IBV isolate (Hopkins & Yoder, 1982; Landman & Feberwee, 2004). Besides synergism with other respiratory agents, for both MG and MS, it was also shown that disease severity may be enhanced due to immunosuppression (Vardaman, Landreth et al., 1973; Giambrone et al., 1977; Morrison et al., 1977) and/or adverse climate and house conditions (Sato et al., 1973; Yoder et al., 1977; Kempf et al., 1988).

Epidemiology

Knowledge about the epidemiology of MG and MS is vital for the design of monitoring and control programmes of both species. It includes knowledge about the transmission routes, survival of MG and MS in the environment, and the occurrence of MG and MS in avian species other than chickens and turkeys. Also, the development of molecular typing methods has contributed to the knowlegde regarding the epidemiology of MG and MS.

Transmission routes

The economic relevance of MG related to respiratory disease, egg production losses, and reduced hatchabililty was already acknowledged in the 1950s (Floyd et al., 1952; Crawley & Fahey, 1955). Subsequently, control of this species became relevant and knowledge with regard to the transmission routes of avian mycoplasmas also became more important. In early studies it was determined that MG and MS infection led to a carrier state and that both species can be transmitted both horizontally and vertically (in ovo), and control of CRD should be based on the control of MG in the reproduction flocks (Fahey & Crawley, 1954b, 1956; Carnaghan, 1961; McMartin, 1968; Vardaman, 1976). Horizontal spread can occur via direct contact and via indirect contact by aerosol transmission or by the introduction of contaminated materials or persons (Fahey & Crawley, 1955a). Furthermore, the study of McMartin and co-workers evidenced the effect of poultry density on the rate of horizontal transmission within a flock (McMartin et al., 1987). Vertical transmission is a major route of spread of the infection, and is of economic relevance regarding international trade. The egg transmission rates are unpredictable and may vary between strains (Olesiuk & Van Roekel, 1960; Carnaghan, 1961; Roberts & McDaniel, 1967; Lin & Kleven, 1982b) and phase of the infection (Olesiuk & Van Roekel, 1960; Carnaghan, 1961; Roberts & McDaniel, 1967; Lin & Kleven, 1982b). The peak egg transmission rates occur about 3–4 weeks after infection and are lower in the chronic phase of the infection (Roberts & McDaniel, 1967). Although the percentage of vertically-infected progeny is low in the chronic phase, horizontal spreading to non-infected chickens still can occur (Fahey & Crawley, 1954b). In some parts of the world, the prevalence of MG has decreased considerably due to the top-down control (Ter Veen et al., 2021).

Survival outside the host

The understanding of reservoirs and survival of both MG and MS outside the host has improved; however, their role in the epidemiology is not yet fully elucidated (Kleven, 1997, 2008; Levisohn & Kleven, 2000; Landman, 2014). Mechanisms of spread between flocks are not yet fully understood, as unexplained outbreaks still occur and may involve direct and indirect transmission by contact, aerosol, and fomites (Fahey & Crawley, 1956; Olsen, Kerr et al., 1964; Anderson & Beard, 1967; Mohammed et al., 1987; Marois et al., 2000, 2005). Also, artificial insemination has been regarded as a route of transmission for MG (Buntz et al., 1986). Multi-age housing and poor hygiene management can be regarded as risk factors for the introduction and transmission of MG and MS. Studies have shown that both mycoplasma species can survive on different materials such as straw, rubber, nose, hair, dust, feed, water, and egg debris, and these materials can be regarded as sources of indirect transmission for both species. However, survival time is low on most materials (1–4 days) except for egg material in which both species can survive for several months (Chandiramani et al., 1966; Christensen et al., 1994; Abolnik & Gouws, 2014). Based on the latter, it can be concluded that materials contaminated with egg debris are a major risk factor in the horizontal transmission of MG and MS. Survival potential outside the host is also evidenced by the identification of biofilm formation by MG. Biofilm formation suggests a survival advantage and increased resistance to disinfectants (Chen et al., 2012; Wang et al., 2017), and is subsequently a risk factor for transmission. A recent study showed that biofilm formation may also increase the survival potency of a pathogen inside the host and, in that way, may also increase the potency for vertical transmission (Chen et al., 2021).

Hosts

Many studies evidenced the presence of MG and MS infections in birds other than chickens and turkeys, including different types of backyard poultry (such as pheasants, partridges, and quails), wild birds (such as sparrows, house and gold finches, pigeons, and flamingos), and waterfowl, such as ducks and geese (Madden et al., 1967; Tripathy et al., 1972; Pascucci et al., 1976; Yamada & Matsuo, 1983; Buntz et al., 1986; Poveda et al., 1986; Ley et al., 1996; Bradbury et al., 2001b; Catania et al., 2016; Michiels et al., 2016). A recent review reported the presence of MG in 56 non-poultry hosts (Sawicka et al., 2020). Experimentally it was also shown that transmission of MG from wildlife to chickens (Stallknecht et al., 1998) could occur only after direct contact (Kleven & Fletcher, 1983; Stallknecht et al., 1998). This latter finding stresses the importance of keeping commercial chicken houses free from wild birds. This also raises the question whether the increasing tendency to keep commercial chickens outdoor increases the risk of introduction of MG and/or MS via contact with wild birds and waterfowl.

Epidemiological conclusions based on molecular tests

In the past decades several molecular typing techniques have been used to unravel the epidemiology of MG and MS. These tests have helped to differentiate between vaccine and field strains, to investigate the role of wild birds in the epidemiology of MG and MS, and to investigate the role of horizontal or vertical transmission in outbreaks of MG or MS. Several studies evidenced the role of molecular epidemiology for the differentiation of vaccine and field strains and identified the persistence or spread of vaccine strains. By the use of Restriction Endonuclease Analysis (REA) MG vaccine strain could be differentiated from field strains (Yogev et al., 1988; Kleven, Browning et al., 1988; Ley et al., 1993). Furthermore, another study, using Random Amplified Polymorphic DNA (RAPD), indicated the persistence of a vaccine strain on a farm. In this study, RAPD patterns similar to that of vaccine strain 6/85 were identified on a farm where this vaccine had previously been used (Charlton et al., 1999). The use of MGc2 and MG IGSR typing showed the spread of live vaccine strain from vaccinated to non-vaccinated flocks in Jordan (Gharaibeh et al., 2011). Two genetic groups (A and B) of MG isolates could be differentiated, and multiple Jordan isolates could not be differentiated from the F-live vaccine strain. Only three isolates could be differentiated as field strains. As the flocks had not been vaccinated and F-live vaccine strain was commonly used in Jordan, the authors suggested it was likely that the F-live vaccine strain had spread from vaccinated to non-vaccinated farms and has, therefore, become the predominant genotype in Jordan. Also, an Egyptian study reported the spread of F-live vaccine strain from vaccinated to non-vaccinated flocks (Khalifa et al., 2014). In another study, the results of molecular genotyping were indicative for the potential spread of ts-11 live vaccine strain and subsequent studies suggested vertical transmission as well as a possible reversion of virulence (El Gazzar et al., 2011). Core genome MultiLocus Sequence Typing (CgMLST) based on 425 conserved genes showed that the ts-11-like isolates, previously obtained from outbreaks in commercial broiler flocks in North-eastern Georgia, were indeed very closely related to the ts-11 vaccine strains (Ghanem et al., 2018). This further suggested that the vaccine was the shared origin of this outbreak in the United States. Furthermore, this study also showed that some of the isolates originating from the UK were very closely related to 6/85 indicating a possible common origin for these isolates.

In another study, high sequence identity based on the MGc2 gene was shown between MG isolates obtained from commercial and backyard turkey flocks in Iran and some of the Iranian turkey isolates and Indian and Pakistan isolates (Rasoulinezhad et al., 2017). This finding indicated the presence of 6/85 vaccine-like strains in the field responsible for clinical outbreaks and which were different from the 6/85 vaccine strain. Discrimination of MS-H live vaccine and field strains by MLST demonstrated the possible spread of MS-H live vaccine from vaccinated to non-vaccinated birds (Dijkman et al., 2017).

Moreover, molecular tests have also been applied to elucidate the role of wild birds in the epidemiology of MG and MS. RAPD has contributed to unravel the epidemiology of outbreaks of MG in free-ranging house finches (Carpodacus mexicanus) which began in the eastern United States and spread throughout the United states, infecting not only house finches but also other songbirds (Fischer et al., 1997; Frasca et al., 1997; Ley, Berkhoff et al., 1997, 2006; Hartup et al., 2000; Mikaelian et al., 2001; Roberts, Nolan & Hill 2001; Roberts, Nolan, Lauerman et al., 2001; Wellehan, Calsamiglia et al., 2001; Wellehan, Zens et al., 2001; Pillai et al., 2003; Cherry et al., 2006). RAPD patterns of MG isolates from songbirds examined from 1994 through 1996 from 11 states, representing three host species, were identical but different from other isolates tested (Ley, Berkhoff et al., 1997). This finding was indicative of a clonal outbreak of MG in songbirds and suggested a single source (Ley, Berkhoff et al., 1997). The previous study also suggested that the outbreaks in finches were not related to MG outbreaks in commercial poultry. RAPD patterns of MG isolates involved in outbreaks of conjunctivitis in evening grosbeaks (Coccothraustes vespertinus) and pine grosbeaks (Pinicola enucleator) during the winter 1998–1999 in Canada were also identical to those of the house finches and gold finches suggesting an on-going epidemic of MG conjunctivitis in eastern North America (Mikaelian et al., 2001). By MLST identical sequence types were identified in backyard poultry and commercial poultry suggesting the potential spread of MG between backyard and commercial poultry (Ter Veen et al., 2021). VlhA sequence analysis was also shown to be not sufficiently discriminatory to distinguish between MS field and MS vaccine strains as was shown by the study of an MS isolate obtained from a captive lesser flamingo (Phoeniconaias minor) in an Italian zoo (Catania et al., 2016). Although no MS-H vaccination was used in the zoo, vlhA typing was unable to distinguish the MS isolate from the MS-H vaccine strain, while adding the sequence analysis of the obg gene suggested a major difference.

Finally, molecular tests have been used to elucidate the source of outbreaks with MG and MS. The use of REA showed that the MG isolate from a peacock was different from the MG isolates obtained from outbreaks in commercial poultry indicating that it was very unlikely that the outbreak in peacocks was the source of the outbreak in commercial poultry (Kleven, Morrow, et al., 1988). REA has also been used for MS (Morrow, Whithear et al., 1990) and was shown to be relevant for epidemiological questions as was demonstrated in an outbreak investigation of MS infection in North Carolina turkeys (Ley & Avakian, 1992) where considerable similarity in REA patterns was observed among MS field isolates suggesting that the outbreaks were related to a common source of infection. RAPD has also been applied by multiple research groups to unravel the epidemiology of MG and MS outbreaks in commercial poultry (Fan et al., 1995; Charlton et al., 1999). RAPD typing indicated that for MG in outbreaks in turkey flocks from different companies and farms in California (Charlton et al., 1999) there was no evidence for spread of MG between companies. It was evidenced by RAPD that a single strain was involved in outbreaks on ranches from the same company. The latter indicated the role of horizontal spread between farms from the same company and underlined the improvement of the hygiene management. RAPD and Pulsed-Field Gel Electrophoresis (PFGE) profiles of MS isolates obtained from 36 layer flocks were very homogeneous (97/99 isolates had similar PFGE and RAPD profiles), suggesting the horizontal transmission of a widely disseminated clone (Dufour-Gesbert et al., 2006). This also complicated determining the precise source of infection to better understand routes of transmission, and indicated that other techniques with higher discriminatory power were needed to further unravel the epidemiology of MS. The 5-gene MLST typing results of 209 MS isolates obtained from 15 different countries, different categories of poultry and farms, and type of lesions (Dijkman et al., 2016) showed different MLST sequence types following geographical origin supporting the hypothesis of regional population evolution. This typing scheme also showed its use for local epidemiological investigation of MS outbreaks. MLST was also used to investigate MG isolates collected from and including the years 2010–2019 and originating from MG outbreaks in Italy (Matucci et al., 2020). MLST clonal complex 1 was the most prevalent clonal complex detected in commercial poultry and also included strains from minor avian species suggesting a possible role of these birds in the evolution of these MG strains.

Monitoring programmes and complicating factors

Monitoring programmes and tests

Well-designed monitoring programmes are essential to be able to control horizontal and vertical spread of MG and/or MS. Although many countries implement a monitoring programme, little information about the contents and efficacy is available in the literature. Examples of successful programmes are the National Poultry Improvement Plan (NPIP) of the US Department of Agriculture (“The National Poultry Improvement Plan ”) for both MG and MS, and the European MG control programme for reproduction stock based on EU Council Directives 90/539/EEC and 2009/158/EC (EU, 1990, 2009). Both US and EU MG control programmes are based on taking a high number of samples starting at the end of the rearing period, frequent sampling in the production period, and culling of infected flocks. For the check of the rearing flocks, the NPIP requires 150 samples per house compared to 60 samples by the EU regulations. During production, the EU requires to take 60 samples per house every 90 days. The NPIP requires to test a total of 75 birds (75 serum samples) every 90 days, or 25 birds (25 serum samples) every 30 days. Breeding companies usually have very strict control programmes to keep all their grandparents and pedigree stock free from both MS and MG. In these programmes, the sampling frequency is usually higher than required by the national programmes, for example every 2–4 weeks.

The first monitoring programmes date from the early 1960s and were based on frequent seromonitoring of small flocks using the tube and rapid plate agglutination test, and the haemagglutination inhibition (HI) test, and frequent attempts to isolate MG from pipped embryos with airsacculitis (Fahey & Crawley, 1954b; Crawley & Fahey, 1955; Beckman et al., 1959). At that time, prevalence was high and there were no vaccines available for MG or MS. Testing of embryos with airsacculitis, using MG PCR or culture, is still performed. Nowadays, monitoring programmes are performed at a flock level and are still based on frequent blood sampling of poultry flocks and testing for the presence of antibodies against MG or MS, and/or the presence of the pathogen itself by culture or PCR (Buim et al., 2009; Kahya et al., 2010; Muhammad et al., 2018).

Studies also showed the potential of egg yolk for monitoring of antibodies at a flock level, but serum is much easier to process in a lab than egg yolk and results in far less biological waste (Mohammed et al., 1986; Brown et al., 1991; Kempf & Gesbert, 1998). Also, testing of day-old chickens (DOC) is performed when the MS or MG status of the breeder flocks is unknown. However, the interpretation of the results can be influenced by many factors, such as serological cross-reactions, antibiotic treatment, MG and/or MS vaccination of the breeder stock, and phase of the infection (Bradbury, 2005; Moronato et al., 2018). For many decades the rapid or serum plate agglutination (RPA or SPA) test and the haemagglutination (HI) test were the standard tests for the monitoring programmes of MS and MG. Difficulties regarding the interpretation of the HI test and RPA test were already recognized in early monitoring programmes (Fahey & Crawley, 1954a). In general, the specificity of RPA and HI tests has been shown to be high; however, several factors have been found that can decrease the specificity (see below). The percentage of RPA and HI non-specific reactions may be reduced by diluting the test serum (Ross et al., 1990; Feberwee et al., 2005, 2020; OIE, 2018). Nowadays, the RPA test is mainly used in monitoring programmes as a screening test as it is quick, relatively inexpensive, and highly sensitive due to its efficient detection of IgM antibodies (Kleven, 1975; Feberwee et al., 2005); MG and MS RPA antigens are still commercially available today. The HI test is less frequently used nowadays as the test is time-consuming, is dependent on the availability of non-commercial reagents, and cross-reactions between antigenically related strains can occur (Goren, 1979; Kleven, Morrow et al., 1988; Kleven et al., 2007; Dingfelder et al., 1991). Commercially available ELISAs are used by many laboratories, as screening test or as a confirmation test. In general, ELISA tests may be slightly less sensitive and more specific than RPA tests, and less specific but more sensitive than HI tests (Avakian et al., 1988; Kaszanyitzky et al., 1994; Kempf et al., 1994; Ewing, Kleven et al., 1996; Kempf & Gesbert, 1998; Feberwee et al., 2005).

Overall, papers show that there is no superior serological test for monitoring programmes; some studies reported that ELISA was not able to detect the antibodies against MG or MS where RPA was able to (Ewing et al., 1998; Kleven et al., 2007). On other occasions, both HI and RPA showed a low sensitivity compared to ELISA (Ewing, Lauerman et al., 1996). Regarding specificity, no serological test has a specificity of 100%. For this reason, it is recommended not to rely completely on only one test system (Feberwee et al., 2005). Further confirmation of serologic results may be by a second antibody test or by detection and identification of MG or MS by isolation or PCR (Luciano et al., 2011; OIE, 2018).

Complications from infections of other Mycoplasma species

Antibodies against heterologous Mycoplasma species that are present in the flock can interfere with serological monitoring for MG or MS, especially in the acute phase of the infection (Kempf et al., 1994; Ewing, Kleven et al., 1996; OIE, 2018; Feberwee et al., 2020). Mycoplasma species other than MG and MS, such as M. pullorum, M. gallinarum, M. gallinaceum, M. iners, M iowae, and M. imitans, are frequently found in commercial and backyard poultry (Bradbury & McClenaghan, 1982; Benčina, Dorrer et al., 1987; Benčina, Mrzel, et al., 1987; Wang et al., 1990; Bradbury et al., 1993, 2001a; Feberwee et al., 2020). Heterologous Mycoplasma species can share some common antigens as shown for MG, MS, and M. imitans using tests such as immunoblots, haemagglutination inhibition (HI) tests, rapid plate agglutination tests, and ELISAs (Bradley et al., 1988; Yogev et al., 1989; Avakian & Kleven, 1990a, 1990b; Bradbury et al., 1993). It has been shown under experimental and field conditions that this decrease of specificity due to these cross-reactions is usually transient and also depends on other factors, such as the antigens used in the test and serum dilution. Using a higher serum dilution can also lower the level of detected cross-reactions due to infections with heterologous Mycoplasma as shown for the RPA and HI test when using the 1:4 or 1:8 dilution as cut-off instead of undiluted or 1:2 diluted serum (Feberwee et al., 2005; OIE, 2018). Doubtful results for MG or MS are best investigated by performing tests with MS antigen (and vice versa) to obtain more information regarding potential cross-reactions (OIE, 2018).

Complications from the recent use of heterologous oil-emulsion vaccines

A transient increase of false-positive reactions in MG and MS serological tests occurs in some flocks recently vaccinated with oil-emulsion vaccines against various agents other than Mycoplasmas (Cullen & Timms, 1972; Glisson et al., 1984; Ahmad et al., 1988; Avakian et al., 1988; Yoder, 1989; Ross et al., 1990). The peak of the number of false-positive reactions is usually a few weeks post-vaccination, after which it gradually disappears in a number of weeks (Glisson et al., 1984). The cause is believed to be the use of products similar to serum components for the production of the Mycoplasma antigen and the production of the antigens for the inactivated vaccine, for example in cell culture or bacteriological culture. Examples of such products are horse serum, foetal calf serum and new-born calf serum (Bradbury & Jordan, 1973; Glisson et al., 1984). When chickens are injected with these antigens in oil emulsions, these also induce antibodies against the remains of the added serum components present in the vaccine. These antibodies are then detected by the similar serum components from the growth medium absorbed to the Mycoplasma cell during in vitro growth (Bradbury & Jordan, 1971, 1972). The use of purified Mycoplasma antigens is expected to decrease the rate of false-positive reactions resulting from antibodies in the test serum directed towards serum components. Using a higher serum dilution can also lower the level of false positives occurring after the recent use of an oil-inactivated vaccine, as shown for the RPA test when using the 1:4 or 1:8 dilution as cut-off instead of undiluted or 1:2 diluted serum (Ross et al., 1990). Another pragmatic and effective approach to distinguish the transient false positives occurring after the recent use of an oil-inactivated vaccine from true positives is to resample the suspected flock a few weeks later and perform the blood tests again. In the case of false positives in the first sampling due to the use of an inactivated vaccine these reactions usually disappear in several weeks.

Complications from antibiotic treatments

Antibiotic treatments that are effective against MG or MS may decrease or delay the antibody response to an infection, especially when the treatment is used in the early stage of the infection (Fahey & Crawley, 1955b; Jordan et al., 1989, 1999; Migaki et al., 1993; Jordan & Horrocks, 1996; Levisohn & Kleven, 2000; Stanley et al., 2001). It will also decrease the number of positives in the PCR and culture.

Complications from infections by low virulence strains

Many comparisons and publications have shown that RPA, HI, and ELISA tests will provide the same diagnosis in a big majority of infected and non-infected flocks. In a minority of cases, RPA, HI and/or ELISA provide conflicting results or all show a lower sensitivity than expected (Kleven et al., 2001), for example in the case of atypical or slow spreading MS and MG strains of low virulence (Yoder, 1986; Dingfelder et al., 1991; Ewing, Lauerman et al., 1996; Kempf et al., 1997). Kleven et al. (2001) reported poor to inconsistent reactivity by RPA, ELISA, and HI-test in MS-infected turkey flocks. Experimental infection of turkeys with one of the isolates (K4463B) either by aerosol or systemically by a combination of intravenous, foot pad, and eyedrop routes showed a clear difference in serological responses. Turkeys challenged by the systemic route responded normally to all serologic tests, whereas those challenged by aerosol either responded very poorly on all serologic tests or were seronegative up to 6 weeks postchallenge, even though they were positive for MS by tracheal culture. These results suggest that turkeys may harbour an upper respiratory infection with MS, while remaining serologically negative. The study of Kempf et al. (1997) also showed that challenge of chickens with a so-called atypical isolate leads to poor serological results and poor mycoplasma culture results.

Complications from the use of live or inactivated MG and MS vaccines

The use of MG or MS vaccines interferes with monitoring programmes. Both live and inactivated vaccines are expected to induce an antibody response. Inactivated MG and MS vaccines are expected to induce a strong serological response. To increase the immunogenicity of the inactivated antigens, immune-potentiating vaccine adjuvants are used (Panigraphy et al., 1981; Barbour & Newman, 1990).

The level of antibodies induced by live conventional vaccines varies and depends on, amongst other factors, the level of attenuation of the strain, way of application, and the interval between vaccination and sampling. In general, F strain-derived live vaccines induce higher levels of antibodies against MG than ts-11-derived vaccines, whereas 6/85-derived live vaccines often induce low or no detectable serological response during the first months post-vaccination (Whithear et al., 1990; Evans & Hafez, 1992; Abd-el-Motelib & Kleven, 1993; Ley, McLaren et al., 1997; Branton et al., 2002; Burnham et al., 2002; Noormohammadi et al., 2002; Noormohammadi & Whithear, 2019). The variation in serological responses to vaccination is underlined by the long-term study of Noormohammadi and Whithear (2019) who reported a slowly increasing serological response to a 6/85 vaccine from 26% at 4 weeks post-vaccination to 100% at 10–36 weeks post-vaccination. In the same study, the initial 100% serological response to the ts-11 vaccine at 4 and 10 weeks post-vaccination dropped to 50–70% positives at 20–36 weeks post-vaccination. These variations in the levels of systemic antibodies after vaccination with live conventional MG or MS vaccines may complicate the diagnosis of a field challenge as seroconversions can also be caused by the variable response in time to the vaccination.

Moronato et al. (2018) studied the serological response of broiler-breeders (vaccinated at week 14 with live vaccine MS-H) at 25, 30, 38, 44, 49, 54, and 59 weeks of age. No field strains were detected during this field study despite intensive monitoring. The birds were located in six barns. On average 79% and 85% of the sera were positive in the RPA and ELISA, respectively. However, there were significant differences in the percentage of positive samples over time and between barns. The difference in percentage of positive ELISA results between the barns at the same age varied from 20% to 100%. The birds of three barns had a consistently high percentage of positives at all ages where as the other barns showed a varying percentage of positives. The authors also report a high variation in ELISA titres.

The use of live vaccines also interferes with the use of culturing and PCR, for example as confirmation tests. Live MG and MS vaccines are expected to be detectable in the respiratory tract for several months to lifelong. In general, less attenuated strains can be detected in a higher percentage of the vaccinated birds and for a longer period (Kleven, 1981; Ley, McLaren et al., 1997; Noormohammadi & Whithear, 2019).

In case MG or MS vaccinated birds are monitored, a certain level of antibodies is already expected meaning that a challenge to a field strain could be detected based on a significant rise in antibody levels (Abd-el-Motelib & Kleven, 1993). For several vaccines, over the last years tests have become available that can differentiate vaccine from field strains, so-called DIVA tests (differentiating infected from vaccinated animals) (Shahid et al., 2014; Kreizinger et al., 2015; Dijkman et al., 2017; Zhu et al., 2017; Sulyok et al., 2019).

An alternative for conventional live MG and MS vaccines are vectorized recombinant vaccines expressing one of several proteins of MG or MS, such as a commercially available recombinant Fowl Pox vector expressing MG 40k and mgc genes (Zhang et al., 2010). The use of such a transient replicating vaccine that induces an immune response only to the proteins expressed from the inserted genes of Mycoplasma would cause hardly or no complications for tests used in monitoring programmes. The detectable serological response for commonly used ELISAs is expected to be very low or even absent (Zhang et al., 2010; Ferguson-Noel et al., 2012). Also, a MG PCR would only be able to detect this vaccine strain during its transient and short replication and only when the primers or probe were designed to detect the insert of the vector vaccine.

Today’s monitoring programmes and issues for consideration

Many developments have taken place since the first monitoring programmes were started. During past decades the worldwide poultry industry has grown considerably and, in parallel, flock sizes and number of birds per farm has also increased. This development increases the importance of a high quality of the diagnosis. Fortunately, the toolbox with diagnostic options and techniques has also grown. The use of the correct diagnostic tests, or combinations thereof, in the various field situations of low to high prevalence, whether or not in combination with vaccinations, requires a great deal of knowledge. Table 1 shows an overview of the use of screening and confirmation tests in monitoring programmes in non-vaccinated and vaccinated flocks.

Table 1. Overview of the use of screening- and confirmation tests in monitoring programmes in non-vaccinated and vaccinated flocks.

			Serology		Detection of pathogen		Conclusion
Vaccinated	Type of vaccine	Infected	Screening test	Confirmation test	Detection of pathogen (screening or confirmation)	DIVA test^a required	Free from infection	Diagnosis of field infection
No	None	No	Negative	Negative (if needed)	Negative (if needed)	No	Negative serology or no pathogen detection
		Yes	Positive	Positive	Positive	No	-	Seroconversion or detection of pathogen
Yes	Inactivated vaccine	No	Positive	Not needed	Negative	No	No further increase of antibody titres, or negative pathogen detection
		Yes	Positive	Not needed	Positive	No		Seroconversion or detection of pathogen
	(also) Live, conventional vaccine	No	Positive	Not needed	Positive	Yes	No further increase of antibody titres, or negative DIVA pathogen detection
		Yes	Positive	Not needed	Positive	Yes		Seroconversion^b or detection of pathogen by DIVA test
	Live, vectored recombinant vaccine	No	Negative	Not needed	Negative	No	Negative serology or no pathogen detection
		Yes	Positive	Positive	Positive	No		Seroconversion or detection or pathogen

^a DIVA (Differentiating Infected from Vaccinated Animals).

^b Note: Several reports show varying levels of antibodies and seroconversions months after vaccination due to response to the vaccine strain. Detection of the field strain is the preferred way for detection for a field infection.

Monitoring of non-vaccinated flocks that are expected to be free of infection is the simplest situation, as any seroconversion or detection of the pathogen indicates an infection. In this situation, false-positive results are a major concern as specificity of tests, certainly all serological tests, is not 100%. Especially when higher sample sizes are used to detect early infections of MG or MS, the risk of false-positive results increases significantly. The use of confirmation tests on sera that tested positive will further increase specificity and may further contribute to a correct decision on the MG or MS status of the flock (Martin et al., 1987). However, a matter of concern is that the use of combined tests to increase the specificity may decrease the sensitivity for the detection of an infection. As only positive sera are retested, a confirmation test can only lower the number of positive samples. This means that the demands of a confirmation test are not only a high specificity but also a high sensitivity. This requirement is especially important in the early phase of infections when some tests might detect the antibody response earlier than others (Avakian et al., 1988; Kempf et al., 1994; Feberwee et al., 2005; OIE, 2018). Confirmation of positive serological tests can also be performed using techniques to detect the bacterium itself, by culturing or, more commonly, by PCR (Ewing, Lauerman et al., 1996). Another way to increase the specificity of serological testing at flock level is to accept a low number or percentage of positive samples before a flock is considered to be positive. OIE (2018) suggests 10% in the RPA MG (when no confirmation test is used). Feberwee et al. (2008) and Ter Veen et al. (2020) accepted one positive sample in the RPA MS using a serum dilution of 1:8.

Flocks that have only been vaccinated with an inactivated vaccine against MG or MS are expected to be positive in serological tests and negative in tests that show the presence of the pathogen. A peak of antibodies is expected shortly after the vaccination, following which the titres will gradually start to decline. A seroconversion during production or the detection of the pathogen will indicate an infection.

Most complicated is monitoring of flocks that have been vaccinated with a live, conventional vaccine. Live MG and MS vaccine strains may persist in the flock and may be detected by general MG and MS PCR tests. Live-vaccinated flocks are also expected to be seropositive to a certain extent, but the percentage of positives and titres vary in time and depend, amongst other factors, on the level of attenuation of the strain, way of application, and the interval between vaccination and sampling (Whithear et al., 1990; Evans & Hafez, 1992; Abd-el-Motelib & Kleven, 1993; Ley, McLaren et al., 1997; Branton et al., 2002; Burnham et al., 2002; Noormohammadi et al., 2002; Moronato et al., 2018; Noormohammadi & Whithear, 2019). Therefore, a DIVA PCR test may give more certainty to monitor field challenge as they are able to discriminate field strains from vaccine strains.

Monitoring of flocks that have only been vaccinated with a live vectored vaccine can be compared with monitoring of non-vaccinated flocks as the vaccine does not persist in the bird and it does not or hardly induce a detectable antibody response. The number of commercially available live vectored vaccines for MG and MS is still very low but that might change in the future due to the strong progress made in this area during the last decade.

To conclude, there is no monitoring programme that is suitable for all conditions; it should be suitable for the actual prevalence and whether flocks have been vaccinated or not. Whatever the situation, it is recommended not to rely solely on one test in monitoring programmes for MG and MS (Feberwee et al., 2005, 2020; Kahya et al., 2010; Luciano et al., 2011; OIE, 2018).

Concluding remarks

The increased knowledge, regarding the expression of clinical disease, epidemiology, vaccination, and diagnostic tools, has allowed further development of ways to control both MG and MS in the commercial poultry industry in the last six decades. Many developments have taken place since the first monitoring programmes were started, including the use of vaccines. Fortunately, the toolbox with diagnostic options and techniques has grown to meet the requirements of the industry. Recognition of new emerging strains and the synergistic effect of other factors in expressing the disease have contributed to the recognition of the clinical and economic importance of MS and MG. The latter has also contributed to the increased urge to unravel the epidemiology of both mycoplasma species to be able to improve the current monitoring and control programmes. There is no monitoring programme that is suitable for all conditions; a monitoring programme has to be suitable for the actual prevalence and whether flocks have been vaccinated or not. Whatever the situation, it is recommended not to rely solely on one test in monitoring programmes for MG and MS. Gaining more information on the expression of clinical disease caused by avian mycoplasma, its epidemiology, and expansion of well-defined collections of isolates from a wide geographical range and period in time are key for filling in the broader epidemiological picture and are crucial for further control of the disease in the commercial poultry industry. Combining all this information has led to novel and significant insights into the origin, diversification, and evolutionary process of avian mycoplasmas both on a local and global scale (Ley et al., 2016; Ball et al., 2018; De la Cruz et al., 2020; Matucci et al., 2020; Sawicka et al., 2020; Ter Veen et al., 2021). The poultry industry is expected to grow in the coming decades to be able to feed the expanding world population. Subsequently, this will lead to an increase in international trade of poultry and eggs, and the possible use of live and or inactivated vaccination programmes. With this background, tight biosecurity and the role of molecular tests with high discriminatory power as tools for unravelling epidemiological questions and knowledge on factors interfering with monitoring programmes are going to play an even more prominent role in the control of both mycoplasma species.

Disclosure statement

No potential conflict of interest was reported by the authors.