Marek's Disease in poultry industry

Marek´s disease (MD), a lymphoproliferative disease caused by Gallid herpesvirus 2, remains a concern for the poultry industry 40 years after the first vaccines were introduced. Although the current situation does not indicate a major problem, the potential for further increases in virulence of MD virus (MDV) remains. In this review I will briefly discuss the current MD situation, problems associated with the proper diagnosis, vaccine failures, current vaccines, and potential future developments.

Gimeno (2004) used a questionnaire to analyze the incidence of MD in 55 countries. Only five countries, including Peru in Latin America, reported economical losses caused by MD since the 1990´s. In the America´s Mexico, Columbia and Venezuela reported outbreaks in the early 2000´s, while the USA, Canada, Brazil and Argentina did not report outbreaks or indicated that the outbreaks during the 1990´s were under control. Unfortunately, the extend of the reported outbreaks was not discussed. In addition to the America´s, outbreaks were reported in a few European countries including Russia with economical losses and China. Recently, John Dunn of ADOL (East-Lansing, MI) conducted a new survey covering responses from 108 countries (personal communication), in which he asked if MD is a problem, if it is frequently diagnosed, and if MD is increasing or decreasing. MD is only occasionally diagnosed in the Americas and the incidence is decreasing in most of South, Central and North America. Exceptions are Bolivia, Columbia, Peru, Uruguay and Venezuela. These responses contrast with the responses to the question if MD is considered to be a problem, with all Latin American countries answering that MD is a considered a problem by most responders. In the rest of the world, MD is occasionally diagnosed, with the exception of several countries in Central Africa and a few countries in East Europe. China, India and Australia provided mixed responses. A remarkable change is the situation in Russia, which reported in 2004 economical losses but the recent survey by Dunn indicated that the frequency is decreasing and that MD is only occasionally diagnosed. Usually MD includes neural lesions, visceral and skin tumors and perhaps transient paralysis in commercial flocks, but until recently there were no reports of early mortality syndrome (EMS) outside of experimental infections. The current interest of consumers in the USA in free-range eggs has led to an increase in backyard flock production systems often using genetically highly susceptible, non-vaccinated Rhode Island Red (RIR) chickens. EMS was diagnosed in a flock of RIR chickens in the absence of other immunosuppressive viruses such as chicken anemia virus or infectious bursal disease virus (Carvallo et al., 2011). An increase in incidence of EMS can be expected in backyard flocks and it will be important to provide proper extension services to this group of producers.

One of the key problems with the questionnaires by Gimeno (2004) and Dunn (personal communication relates to a) was the diagnosis correct and b) what constitutes a break? The proper diagnosis of MD is in principle rather simple, but can be very complex in practice (Witter and Schat, 2003). The American Association of Avian Pathologists recognized this problem and published a tumor diagnosis manual (Witter et al., 2010). Proper diagnosis of MD is in general fairly straight forward in chickens younger than 14 weeks of age, although non-bursal tumors caused by reticuloendotheliosis virus (REV) may need to be excluded. Fortunately, REV nonbursal tumors have not been reported in commercial flocks younger than 14 weeks of age. Lymphoid leukosis is the most likely diagnosis in older birds with bursal tumors, although MD virus and REV can also cause bursal tumors. In the past MD was associated with younger birds, but vaccination breaks do occur in older birds (Witter, 2001). If tumors are present in older birds in the absence of bursal tumors, MD is the default diagnosis. The detection of virus by conventional virus isolation, or by advanced techniques such as real-time quantitative (q)PCR or by loop-mediated amplification (LAMP) (Wozniakowski et al., In press), can show the presence of MDV in blood lymphocytes, feather follicle epithelium (FFE), and feather pulp. Gimeno et al. (2008) suggested that qPCR may be helpful for the identification of birds at risk of developing tumors based on the level of viral DNA copies. In a subsequent paper it was suggested that virus replication loads could be used to divide birds in virus-negative, latent infection and tumor bearing groups (Gimeno et al., 2011). In order to validate this approach a considerable amount of additional work is needed to determine true cut-off values for the discrimination between tumor and non-tumor bearing birds. A potential problem with this approach will be temporal reactivation of MDV replication without tumor development but resulting in a spike of MDV genome copies. Immunohistochemistry (IHC) or FACS analysis using (monoclonal) antibodies to lymphocyte markers are routine techniques and can be used to characterize tumors as B cell or T cell tumors. However these techniques require properly prepared tissues for IHC or freshly obtained individual tumor cells for FACS to be useful. If properly fixed tissues are available, IHC can be used to further characterize tumors. For example, the presence of meq protein, the MD oncogene protein expressed in all MD tumors, is the best way to positively diagnose MD especially if pp38, a MDV protein expressed during cytolytic infection, is absent or only present in a few of the Meq-positive cells. Likewise, IHC can be used to demonstrate the presence of REV and avian leukosis virus (ALV) proteins in tumors.

The finding that MD breaks occur in older birds raises the interesting question if these late tumors are caused by reactivation of early infections (old infection hypothesis) or the consequence of de novo infections with more virulent strains (new infection hypothesis) (Witter, 2001). Witter and Gimeno (2006) and Ikezawa et al. (2010) infected SPF or vaccinated commercial chickens after 18 weeks of age. Although the two groups used different protocols, their findings were in general agreement. Birds not previously vaccinated or exposed to MDV remained fully susceptible to MD, while birds previously exposed to MD or vaccinated remained largely free of MD suggesting that late breaks were probably caused by reactivation thus supporting the old infection hypothesis. The problem with these studies was the inability to differentiate between the virus used to infect at a young age and the challenge virus used later in life. To further address the question what the consequences are of superinfection, Dunn et al. (2010) used Md5, a very virulent (vv)MDV, in which the pp38 gene of CVI988 was ed (Md5//38CVI). The pp38 gene of CVI988 has a single nucleotide polymorphism resulting in a difference of one amino acid with pp38 from oncogenic MDV strains. This difference can be detected by IHC and pyrosequencing. Challenge with the virulent JM strain or Md5 followed by Md5//38CVI or vice versa 2 days later resulted in a high incidence of dual infections. If the second challenge was given after 14 days the superinfection was significantly reduced but not totally eliminated. These results also suggest that the old infection hypothesis may be a more likely explanation for late breaks.

In general the current vaccines provide excellent protection against MD as is evident from the data on the current situation (Dunn, personal communication) and the extremely low condemnation rates of broilers in the USA (Agricultural Statistics Board USDA, 2011). However, breaks do occur from time to time raising the specter of a new wave of even more virulent strains than the vv+ strains reported by Witter (1997). In many instances the breaks are the consequence of industry practices especially in the broiler industry, where vaccine doses are routinely used below the recommended levels, down-time between flocks may be as short as 14 days, and litter is frequently re-used for a number of cycles (Schat and Baranowski, 2007). The negative impact of diluting MD vaccines on protective immunity was recently reemphasized in an elegant study by Gimeno et al (2011). They used qPCR to measure vaccine replication in feather pulp. Dilution of vaccines led not only to a decrease in virus levels in the feather pulp and decreased protection especially after challenge with vv+ strains of MDV, but also to lower body weights even after challenge with Md5. The decrease in body weights was especially marked in female birds. The use of re-used litter or placing new birds in insufficiently cleaned houses adds to the risks of breaks. Protection levels in HVT-vaccinated birds increased from 40% to 80% between 2 and 7 days post vaccination when exposed to MDV-containing dust, (Fakhrul Islam et al., 2008). The increase in protective immunity makes sense in view of the development of protective immunity which starts with the innate responses around 3 days post vaccination followed by acquired, cell-mediated immunity between 6 and 7 days post vaccination (Schat and Markowski-Grimsrud, 2001).

The analysis of MD breaks is often complicated and not necessarily caused by new strains with increased pathogenicity. MD breaks can be the consequence of infection with other pathogens especially chicken anemia virus (CAV), which has been linked to many MD breaks (Schat, 2004). The main reason for the impact of CAV on the development of MD is the replication of CAV in dividing cytotoxic T lymphocytes (CTL) resulting in apoptosis of the CTL and decreased cell-mediated immunity (CMI) (Markowski-Grimsrud and Schat, 2003). Recently, Haridy et al. (2009) reported severe depletion of CD8+ cells in spleen and thymus of chickens infected with MDV at one day of age and CAV at 4 weeks of age further demonstrating the impact of CAV on CMI responses to MDV.

Thus far there is limited evidence for the appearance of new more virulent MDV strains than the current vv+ strains. A deletion in the gL gene was linked to decreased protection (Tavlarides-Hontz et al., 2009), which is of interest because Markowski-Grimsrud and Schat (2002) detected CTL against gL in the resistant N2a chicken line. It is not known, however, if the mutant gL protein can be detected by CTL. Predominant field strains in China were different than CVI988 for gI, gE and meq genes. The change in gI is especially interesting because gI has been linked to protective immunity (Lee et al., 2004) and the generation of specific CTL (Markowski-Grimsrud and Schat, 2002). MDV isolated from MD breaks in Poland had ions of LTR-REV sequences and ions in the meq gene (Woźniakowski et al., 2011). None of these new strains have been evaluated for increased pathogenicity using the method outlined by Witter et al (2005) and is therefore not clear if these strains are indeed more pathogenic.

Monitoring of vaccine-induced protection has been problematic and required virus isolation in cell culture to demonstrate viremia. The rapid acceptance of qPCR has made it possible to use feather pulp or feather tip cells (FTC) to analyze levels of vaccine viruses. Baigent et al (2005; 2006) first established that CVI988 can be detected in FFE starting at 7 days post vaccination and that samples taken at 21 days post vaccination provided optimal samples. Subsequently, they established that approximately 132 genome copies/10.000 FTC are correlated with 90% protection (Baigent et al., 2007). More recently, they developed a method to differentiate field strains of MDV from CVI988 using a BAC-cloned CVI988 vaccine. Primers for U_S2 were used for the detection of field virus and primers for the BAC sequence replacing U_S2 in the vaccine strain. CVI988 reduced virus levels of RB-1B significantly in spleen, kidney, liver, and FTC compared to levels in nonvaccinated controls (Baigent et al., 2010). Although the latter approach cannot be used in commercial situations, the use of FTC or feather pulp to monitor the vaccination status of flocks is expected to become standard practice especially to retrospectively analyze breaks. Storage of feather pulp DNA on FTA^® cards providing an easy storage system will further facilitate such an approach (Cortes et al., 2009).

World-wide there is an increase in revaccination protocols especially in breeders and to some degree in layers (Dunn, personal communication 2011). The rationale for revaccination does not (yet) have a scientific basis and the older literature does not provide evidence for improved protection (Witter, 2001). Wu et al. (2009) showed improved protection in field trails when birds were vaccinated at 1 day of age with CVI988 or HVT followed by a second vaccination at 7 days using the same vaccines. In laboratory experiments they found that double vaccination resulted in two cytolytic replication phases of the vaccine virus. Challenge with RB-1B in revaccinated birds resulted in a single cytolytic infection phase of RB-1B, while two cytolytic replication cycles were observed in single vaccinated birds, perhaps providing a model to study the mechanisms behind improved protection by revaccination. On the other hand, Le Galludec (2008) using qPCR assays and feathers found that revaccination at day 7 interfered with vaccine replication and that revaccination was not recommended.

The big question is what will happen if vv++(+) strains do appear. Currently CVI988 is the "gold standard" for MDV vaccines and it may be very difficult to find a better vaccine strain than CVI988 (Witter and Kreager, 2004). Recently, several papers have been published showing that deletion of meq from the Md5 strain protects better than CVI988 (e.g., Lee et al., 2010). However, during the 5^th International Workshop on the Molecular Pathogenesis in October 2010 (Athens, GA) it was mentioned that rMd5∆meq caused severe thymus and bursa atrophy in maternal antibody negative chicks preventing licensing in the USA under current regulations. Other approaches such as vaccination with HVT combined with plasmid DNA expressing chicken interferon-γ have shown some promises (Haq et al., 2011), but the general experience with using plasmid DNA expressing cytokines or recombinant vaccines expressing cytokines has been disappointing.

The recent finding that small RNA fragments can interfere with transcription and translation (RNAi or RNA interference) has led to a new field in developing protection against disease. Insertion of small DNA fragments coding for small hairpin RNA sequences in the germ-line of plants and animals leads to the formation of small double-stranded (ds)RNA fragments of less than 30 nucleotides. Through a complex enzymatic process the dsRNA is separated and the antisense strand can bind to mRNA resulting in the destruction of the mRNA. Selection of appropriate sequences complementary to unique sequences in pathogens will prevent the replication of pathogens. Chen et al. (2009) using a lentivirus vector generated transgenic chickens expressing RNAi sequences against gB and ICP4. Challenge of the transgenic birds resulted in a decrease in viremia levels and a reduced disease incidence. It is too early at this time to know if this approach will work sufficiently well in combination with HVT or HVT/SB-1 vaccines to improve protection significantly. Serotype 1 vaccines cannot be used in these transgenic birds, because the RNAi sequences will interfere with replication of these vaccines. It is also not clear if the consumer will accept transgenic chickens. Insertion of RNAi sequences in HVT has also been reported resulting in moderate reduction in viremia of the challenge virus (Lambeth et al., 2009). The problem with rHVT-RNAi is that the interference will only work in infected cells expressing the RNAi sequences, which will leave a very large number of nonprotected lymphocytes available for the challenge virus.

The recent explosion of HVT recombinant vaccines expressing genes for other pathogens has been a mixed blessing. Each individual recombinant provides in general strong protection against the specific pathogen, however using more than one recombinant reduces the protection against some of the pathogens. This phenomenon has been discussed at several meetings but to my knowledge no peer-reviewed papers have addressed this issue. At this time it is not clear if this is simply a problem of different passage levels of the recombinant virus with different replication efficiencies or that there is a fundamental problem using different recombinant vaccines at the same time.

In conclusion, the MD situation seems to be reasonably controlled with little evidence for the appearance of new more virulent strains. Vaccine protection can be improved by using the recommended doses of vaccine in combination with sound biosecurity measures and perhaps revaccination in problem areas. If this is needed I feel that the second vaccination needs to be given as soon as possible, perhaps upon arrival at the farm rather than at or after 6 days of age. Technology advances such as qPRC will allow improved monitoring of vaccine-induced protection and perhaps a better understanding of reasons why breaks occur. Unfortunately, there are currently no new vaccines ready to replace CVI988 if more virulent MDV strains appear.

Agricultural Statistics Board, USDA, data released June 1, 2011.

Baigent SJ, Petherbridge LJ, Howes K, Smith LP, Currie RJ & Nair VK. 2005. Absolute quantitation of Marek´s disease virus genome copy number in chicken feather and lymphocyte samples using real-time PCR. J Virol Methods 123:53-64.

Baigent SJ, Smith LP, Nair VK & Currie RJ. 2006. Vaccinal control of Marek´s disease: current challenges, and future strategies to maximize protection. Vet Immunol Immunopathol 112:78-86.

Baigent SJ, Smith LP, Currie RJ & Nair VK. 2007. Correlation of Marek´s disease herpesvirus vaccine virus genome load in feather tips with protection, using an experimental challenge model. Avian Pathol 36:467-474.

Baigent SJ, Smith LP, Petherbridge LJ & Nair VK. 2010. Differential quantification of cloned CVI988 vaccine strain and virulent RB-1B strain of Marek´s disease viruses in chicken tissues, using real-time PCR. Res Vet Sci. DOI: 10.1016/j.rvsc.2010.08.002.

Carvallo FR, French RA, Gilbert-Marcheterre K, Risatti G, Dunn JR, Forster F, Kiupel M & Smyth JA. 2011. Mortality of one-week-old chickens during naturally occurring Marek´s disease virus infection. Vet Pathol Online DOI: 10.1177/03009810395727.

Chen M, Payne WS, Dunn JR, Chang S, Zhang HM, Hunt HD & Dodgson JB. 2009. Retroviral delivery of RNA interference against Marek´s disease virus in vivo. Poult Sci 88:1373-1380.

Cortes AL, Montiel ER, & Gimeno IM. 2009. Validation of Marek´s disease diagnosis and monitoring of Marek´s disease vaccines from samples collected in FTA cards Avian Dis 53:510-516.

Dunn JR, Witter RL, Silva RF, Lee LF, Finlay J, Marker BA, Kaneene JB, Fulton RM & Fitzgerald SD. 2010. The effect of the time interval between exposures on the susceptibility of chickens to superinfection with Marek´s disease virus. Avian Dis 54:1038-1049.

Fakhrul Islam AF, Walkden-Brown SW, Groves PJ & Underwood GJ. 2008. Kinetics of Marek´s disease virus (MDV) infection in broiler chickens. 1: Effect of varying vaccination to challenge interval on vaccinal protection and load of MDV and herpesvirus of turkey in the spleen and feather dander over time. Avian Pathol 37:225-235.

Gimeno IM. 2004. Future strategies for controlling Marek´s disease, pp. 186-199. In Marek´s disease: en evolving disease, Davison F, Nair V (eds.). Elsevier Academic Press, London.

Gimeno IM, Cortes AL, & Silva RF. 2008. Load of challenge Marek´s disease virus DNA in blood as a criterion for early diagnosis of Marek´s disease tumors. Avian Dis 52:203-208.

Gimeno IM, Cortes AL, Montiel ER, Lemiere S & Pandiri AKR. 2011. Effect of diluting Marek´s disease vaccines on the outcomes of Marek´s disease virus infection when challenged with highly virulent Marek´s disease viruses. Avian Dis 55:263-272.

Haq K, Elawadli I, Parvizi P, Mallick AI, Behboudi S & Sharif S. 2011. Interferon-γ influences immunity elicited by vaccines against very virulent Marek´s disease virus. Antiviral Res 90:218-226.

Haridy M, Goryo M. Sasaki J & Okada K. 2009. Pathological and immunohistochemical study of chickens with co-infection of Marek´s disease virus and chicken anemia virus. Avian Pathol 38:469-483.

Ikezawa M, Goryo M, Sasaki J, Haridy M. & Okada K. 2010. Late Marek´s disease in adult chickens inoculated with virulent Marek´s disease virus. J Vet Med Sci 72:1539-1545.

Lambeth LS, Zhao Y, Smith LP, Kgosana L & Nair V. 2009. Targeting Marek´s disease virus by    RNA interference delivered from a herpesvirus vaccine. Vaccine 27:298-306.

Le Galludec H. 2008. Clinical application of diagnostics (PCR) in the management of Marek´s disease: The ultimate goals. In: Marek´s disease symposium CP Group, Chengdu, China.

Lee LF. Bacon LD. Yoshida S, Yanagida N. Zhang HM & Witter RL. 2004. The efficacy of recombinant fowlpox vaccine protection against Marek´s disease: Its dependence on chicken line and B haplotype. Avian Dis 48:129-137.

Lee LF, Kreager KS, Arango J, Paraguassu A, Beckman B, Zhang HM, Fadly A, Lupiani B & Reddy SM. 2010. Comparative evaluation of vaccine efficacy of recombinant Marek´s disease virus vaccine lacking Meq oncogene in commercial chickens. Vaccine 28:1294-1299.

Markowski-Grimsrud CJ & Schat KA. 2002. Cytotoxic T lymphocyte responses to Marek´s disease herpesvirus-encoded glycoproteins. Vet Immunol Immunopath 90:133-144.

Markowski-Grimsrud CJ & Schat KA. 2003. Infection with chicken anemia virus impairs the generation of antigen-specific cytotoxic T lymphocytes. Immunology 109:283-294.

Schat KA & Markowski-Grimsrud CJ. 2001. Immune responses to Marek´s disease virus infection. Curr Top Microbiol Immunol 255:91-120.

Schat KA. 2004. Marek´s disease immunosuppression, pp. 142-155. In Marek´s disease: an evolving disease, Davison F, Nair V (eds.). Elsevier Academic Press, London.

Schat KA & Baranowski E. 2007. Animal vaccination and the evolution of viral pathogens. Rev Sci Tech Off Int Epiz 26:327-338.

Tavlarides-Hontz P, Kumar PM, Amortegui JR, Osterrieder N & Parcells MS. 2009. A deletion within glycoprotein L of Marek´s disease virus (MDV) field isolates correlates with a decrease in bivalent MDV vaccine efficacy in contact-exposed chickens. Avian Dis 53:287-296.

Witter RL. 1997 Increased virulence of Marek´s disease virus field isolates. Avian Dis 41:149-163.

Witter RL. 2001. Protective efficacy of Marek´s disease vaccines. Curr Top Microbiol Immunol 255:57-90.

Witter RL & Schat KA. 2003. Marek´s disease, pp 407-464. In Diseases of Poultry Saif YM, Barnes HJ, Glisson JR, Fadly AM, McDougald LR, Swayne DE (Eds.) Diseases of Poultry. Iowa State University Press, Ames, IA.

Witter RL, & Kreager KS. 2004. Serotype 1 viruses modified by backpassage or ional mutagenesis: Approaching the threshold of vaccine efficacy in Marek´s disease. Avian Dis 48: 768-782.

Witter RL, Calnek BW, Buscaglia C, Gimeno IM & Schat KA. 2005. Classification of Marek´s disease viruses according to pathotype: philosophy and methodology. Avian Pathol 34:75-90.

Witter RL & Gimeno I. 2006. Susceptibility of adult chickens, with and without prior vaccination, to challenge with Marek´s disease virus. Avian Dis 50:354-365.

Witter RL, Gimeno IM, Pandiri AR & Fadly AM. 2010. Tumor diagnosis manual: The differential diagnosis of lymphoid and myeloid tumors in the chicken. American Association of Avian Pathologists, Jacksonville, FL.

Woźniakowski G, Samorek-Salamonowicz E & Kozdruń W. 2011. Molecular characteristics of Polish field strains of Marek´s disease herpesvirus isolated from vaccinated chickens. Acta Vet Scand 53:10-17.

Woźniakowski G, Samorek-Salamonowicz E & Kozdruń W. In Press, Rapid detection of Marek´s disease virus in feather follicles by loop-mediated amplification (LAMP). Avian Dis.

Wu C, Gan J, Jin Q, Chen C, Liang P, Wu Y, Liu X, Ma L & Davison F. 2009. Revaccination with Marek´s disease vaccines induces productive infection and superior immunity. Clin Vaccine Immunol 16:184-193.