Correlations of genetic resistance of chickens to Marek's disease viruses with vaccination protection and in vivo response to phytohemaglutinin

Twenty-three genetic groups of experimental and commercial meat and egg chickens were injected with moderately virulent BC-1 (exp 1) or highly virulent RB-LB (exp 2) Marek's disease (MD) virus. Birds of 7 genetic groups were divided into vaccinated and non-vaccinated groups and exposed by contact to the virulent RB-1B virus in exp 3. Response to phytohemaglutinin (PHA) injected in wing webs was measured in adult birds of all 23 groups (exp 4) to assess its relationship to MD resistance. There was a high correlation (0.8) between resistance of the genetic groups to the two viruses indicating that selection for resistance to one virus would be expected to improve resistance to the other virus. Regression of MD incidence in vaccinated birds on that in non-vaccinated birds resulted in regression coefficients of 0.41, 0.23, and 0.31% for males, females and combined sexes respectively, indicating that MD incidence increased linearly in vaccinated birds in relation to their genetic susceptibility to MD. Two significant correlations in males suggested that high swelling response to PHA may under some conditions be associated with MD resistance. However, the correlation coefficients were inconsistent and it was concluded that swelling response to PHA inoculated in the wing web is not predictive of MD resistance. chicken / Marek's disease / vaccination / phytohemaglutinin R.ésumé-Corrélations entre la résistance génétique de poulets aux virus de la maladie de Marek, la protection de la vaccination et la réaction in vivo à la phytohémagglutinine. Vingt-trois types génétiques de poulets, représentant des souches «chair» et «ponte» expérimentales et commerciales, ont été inoculés avec deux virus de la Maladie de Marek: le virus BC-1, modérément virulement (expérience 1) et le virus * Animal Research Centre Contribution No 1652 RB-1B, fortement virulent (expérience 2). Les sujets de sept types génétiques ont été repartis en groupes vaccinés et non vaccinés et ont été exposés par contact au virus virulent RB-1B dans l'expérience 3. La réaction à la phytohémagglutinine (PHA) injectée dans les membranes alaires a été mesurée chez des sujets adultes des 23 types génétiques (expérience 4) pour évaluer sa liaison avec la résistance à la maladie de Mareck. On constate une forte corrélation (0.8) entre la résistance des types génétiques aux deux virus, ce qui montre que la sélection pour la résistance à un virus semble améliorer la résistance à l'autre virus. La régression de la fréquence de la maladie de Marek chez les sujets vaccinés sur …


INTRODUCTION
Marek's disease (MD) is caused by a herpes virus that induces neoplastic transformation of host T-cells, resulting in formation of lymphoid tumors. Protection by vaccines is not complete and the combination of both vaccination and genetic resistance is required for optimum protection , Gavora and Spencer, 1979). Appearance of very virulent strains of MD virus associated with increased MD losses in vaccinated flocks (Witter, 1988) emphasizes the need to improve vaccines and to increase levels of genetic resistance.
In this context, questions of practical importance are (1) whether genotypes resistant to moderately virulent MD viruses are also resistant to highly virulent viruses, and (2) what is the degree of protection by vaccination against the virulent viruses in genotypes that differ in their natural MD resistance. Genetic improvement of resistance can be accomplished by direct selection based on response to MD virus or, more desirably, on marker traits measurable without exposure to the pathogen. Response of chickens to phytohemaglutinin (PHA) was considered a potential marker trait for this purpose. T-cells play a dual role in the pathogenesis of MD, in that they are both the target cells for neoplastic transformation, and act with natural killer cells, in defence against MD tumors (Sharma et al, 1977, Sharma, 1981. Susceptibility to MD tumors may be linked to strong cell-mediated immune response and is influenced by both age and genotype of the bird (Calneck, 1986) and MD resistance is, at least partly, the property of the target T-cells (Gallatin and Longenecker, 1979).
Response of chickens to PHA injected intradermally is a measure of cell-mediated immunity involving T-cells (Goto et al, 1978), although the response is cellularly heterogeneous (Edelman et al, 1986). Response of chickens to PHA differs among commercial stocks (Van der Zijpp, 1983), or experimental lines (Lamont and Smyth, 1984) and is influenced by both sex and major histocompatibility haplotype (Taylor et al, 1987).
The relationship of PHA response and MD resistance is not clearly understood. A line of chickens selected for high plasma corticosterone had an impaired in vitro response of lymphocytes to PHA and greater MD tumor incidence and mortality than a low corticosterone line (Thompson et al, 1980). In contrast, Lee and Bacon (1983) reported that increased in vitro response of lymphocytes to phytohemaglutinin was associated with increased susceptibility to MD. However, Calnek et al (1989) dit not observe any general correlation between the responses of multiple genetic groups of chickens to mitogens Concavalin A and PHA or mixed lymphocyte reaction, and MD susceptibility.
In this study, correlations between resistance of several genetic groups of chickens to two strains of MD virus that differed widely in virulence were investigated and the relationship between genetic resistance to MD and protection by vaccination was assessed. The relationship of genetic resistance to MD with swelling response induced by the injection of PHA into the wing web is also reported.

Chickens
A description of the genetic groups used in the study is given in table I and the populations used are shown in figure 1. The parental populations were reared intermingled in floor pens and housed in individual cages as adults. They were vaccinated for MD, infectious bronchitis and Newcastle disease, as well as avian encephalomyelitis, and were fed mash rations throughout their lifetime. They were given a uniform light diet in all generations. No major disease outbreak was experienced in any of the parental flocks or the 1983 flock used for the PHA test. In all these flocks, rearing mortality was less than 8% and laying house mortality less than 10%. In addition to the above parents, parallel specific pathogen-free (SPF) parent populations for genetic groups CS, CK, and NH were maintained on a filtered-air, positive-pressure building where they received no vaccines and were free of Marek's disease virus and other avian pathogens.
Marek's disease challenge tests (exp 1, 2 and 3) For the MD challenge tests the birds were in floor pens in an isolation facility . At 3 weeks of age, each bird was inoculated intraperitoneally with the respective MD virus isolant. In exp 1, the inoculum contained the BC-1 virus  and the birds produced from both conventionally housed and SPF parents were observed for 63 d after inoculation. In exp 2 the inoculum was the RB-1B virus (Schat et al, 1981) and the duration was 56 d after inoculation.
The inocula for both experiments were from lots of cell-associated virus stored in liquid nitrogen that had previously been tested for pathogenicity.
Exp 3 included 130 to 147 birds from each of 7 genetic groups (A,3R,7,8,8R,CS, and CK). Approximately half of these birds were vaccinated on the day of hatch with 6000 plaque-forming units of cell-associated herpes virus of turkeys. The birds were kept in isolation until 14 d of age and were then exposed to seeder birds previously infected with the RB-1B virus. The birds were killed at 53 d after the exposure.
All birds that died or were killed because of illness, and survivors that were killed at termination of the tests, were necropsied. MD incidence was based on gross lesions.

Phytohemaglutinin (PHA) response test (Exp 4)
The dose and inoculation site for this experiment was determined on the basis of a preliminary test using 30 adult White Leghorn females and 12 adult White Leghorn males inoculated with 75, 500 and 1250 mg of PHA per bird in the wattle or wingweb. Wing webs were found easier to measure as wattles tended to be soft and pliable. Of the doses tested, 500 and 1250 mg gave similar swelling responses in the wing web.
The procedure used in exp 4 was similar to that of Van der Zijpp (1983). Adult birds (482 d of age at PHA inoculation) were each inoculated intradermally with 0.125 ml of phosphate buffered saline (PBS) in the left wing web. The same volume of PBS containing 500 mg PHA * was inoculated in the right wing web. Prior to the inoculation the wing webs were plucked free of feathers and the inoculation sites, close to but not on the edge of the web were marked with a felt pen.
Thickness of the wing web was measured before inoculation, as well as 24 and 48 h after inoculation, using Miluyo electronic micrometer, model No 293-701 that applied constant pressure on all wing webs, independent of the operator. The swelling index (I) was calculated as where &dquo;T&dquo; is the thickness and subscripts &dquo;R&dquo; and &dquo;L&dquo; indicate the right and left wing web and &dquo;1&dquo; and &dquo;2&dquo; indicate thickness before and after inoculation, respectively.

Statistical analyses
Spearman's rank and Pearson's product-moment correlation between resistance to the BC-1 and RB-1B viruses and PHA response were calculated on the basis of genetic group means. The dependence of MD incidence in vaccinated birds on MD incidence in non-vaccinated birds was assessed by linear regression using genetic group means. MD incidence among genetic groups and vaccination treatments was compared by homogeneity X 2 tests and differences between grouping of genetic groups by the student's t-test. Individual bird swelling index data after PHA challenge were subjected to analysis of variance using a model containing the effects of genetic group, sex and their interaction.

RESULTS
Resistance to the BC and RB-LB isolants of MD virus MD incidence after challenge with the BC-1 and the RB-1B isolants of MD virus differed widely in exp 1 and 2 (table II). In exp 1, the overall MD incidence induced by BC-1 was close to 15% and this was significantly lower (P < 0.01) than the 47% MD incidence induced by RB-1B in exp 2.
The ranges of MD incidence among the genetic groups tested were from 0 to 46.4% in males and 0 to 89.7% in females in exp 1, and from 5.7 to 93.7% in males and from 4.1 to 97.2% in females in exp 2. In genetic groups 8R, XP02, and XP21 there was no incidence of MD after BC-1 challenge while MD was observed in all genetic groups after challenge with RB-LB (table II). Among the birds challenged with BC-1 there was a significant sex difference (P < 0.05), the incidence being 11.5% higher in females than in males. The corresponding sex difference of 4.2% in birds challenged with RB-1B was not statistically significant. With the exception of strain NH females, MD incidence in strains CS, CK, and NH in exp 1 was not significantly different between birds produced from conventionally and SPF housed dams.
The relationship between resistance of the genetic groups to the BC-1 and RB-1B challenge was expressed in terms of Spearman's rank and Pearson's productmoment correlations (table III). the correlations for males and females separately, as well as for sexes combined were all high and significant, although correlations tended to be higher in females.

Relationship of genetic resistance to MD and vaccination protection
Incidence of Marek's disease in the non-vaccinated birds exposed by contact to RB-LB in exp 3 at 2 weeks of age (fig 2) was in good agreement with that in the same genetic groups challenged by injection at 3 weeks of age in exp 2 (table II), although the average MD incidence in contact challenge was 6.6% lower. Vaccination conferred significant protection (P < 0.05) to both males and females and there were high and significant correlations between MD incidence in the vaccinated and non vaccinated birds (table III). Linear regressions were fitted for the relationship between MD incidence in nonvaccinated birds. The regression accounted for 92, 68 and 95% of total variation (R 2 ) for males, females and sexes combined. The respective regression coefficients were 0.41, 0.23, and 0.31 for males, females and sexes combined. Thus, in this experiment, MD incidence in vaccinated birds increased linearly with their genetic susceptibility ( fig 2) and a combination of genetic resistance with vaccination resulted in the best protection.
Marek's disease resistance and response to phytohemaglutinin The means of the swelling index measured at 24 h after injection with PHA are shown in table II. The mean swelling at 48 h post injection and the differences between the 48 h and 24 h swelling were also calculated but are not shown in the table. Although the time of peak response was not determined, the 24 h response comparisons between sexes, the greater swelling was in males in 11 genetic groups and in females in 12 genetic groups. The tendency was for swelling to develop more slowly and to peak later in females than in males.
Rank-order and product-moment correlations of MD incidence and wing web swelling after PHA inoculation are shown in table V. Only the negative correlations of the 48 h swelling index of males with MD resistance reached statistical significance. The remaining correlations of 24 h and 48 h swelling indices for males were also negative, while those for females were inconsistent in both sign and magnitude.

DISCUSSION
Incidence of Marek's disease among the genetic groups in both exp 1 and 2 varied widely and in exp 2 spanned almost the entire range from 0 to 100% (table II). The incidence of MD lesions was described in more detail by Spencer et al (1984) showing that the predominant lesion in BC-1 inoculated females was in the ovary and in males in nerves. The RB-1B virus induced a high proportion of visceral lesions in both males and females. As observed previously , BC-1 induced a higher incidence of MD in females than males. No significant sex difference was observed in MD incidence after RB-1B challenge. This, together with the large difference between the overall MD incidences after BC-1 and RB-1B challenge, emphazises the genetic divergence between the two virus strains. The correlations between resistance to the MD viruses (table III), as well as that between this resistance and PHA (table V), express genetic relationships because the correlations are based on means of the genetic groups, which are themselves genetic characteristics.
Since there were high correlations between resistance to BC-1 and RB-1B, selection for resistance to one of the viruses would also be expected to improve resistance to the other. This conclusion was supported by comparisons of related genetic groups that have undergone various degrees of selection. There were 3 such sets of genetics groups of Leghorns, each originating from a different genetic base population (table I). The sets consisted each of a strain selected for high egg production, egg weight and related economically important traits including viability (P-strains), and of strains selected for resistance to the BC-1 virus in addition to the above traits (R-strains). Selection for resistance to the BC-1 significantly (P <_ 0.01) improved resistance of the R-strains 3R and the 8R to the RB-1B virus compared with their respective P-strain counterpart strains 1 and 8 (table II). In the third set, there was no significant difference between the P-strain 2 and R-strain 2R in RB-1B resistance.
Our results from lines XP02 and XP21 confirm the resistance of the B2i haplotype (XP21) and B 2 haplotype (XP02) against multiple MD viral strains (Gavora et al, 1986;Bacon 1987). The commercial stocks differed widely in MD resistance: Leghorn stock A was among the most resistant group, however, stocks B and C were far more susceptible (table II).
The linear relationship between genetic resistance to MD and protection by vaccination (fig 2) was based on only 7 genetic groups but should have a broad validity for the virus strain and vaccine type tested, as the resistance of the nonvaccinated groups practically spanned 0 to 100%. The positive correlation between resistance to the RB-LB and BC-1 isolants of MD virus suggest that the relationship of MD resistance and protection by vaccination may also be valid for other MD viruses.
In the determination of the PHA swelling index, possible differences in wing web thickness of the right and left wing or the variation in the response to PBS were not considered as it was shown that they are negligible (Van der Zijpp, 1983).
Overall, the levels of response to PHA in this study were lower than those observed by Van der Zijpp (1983) and Lamont and Smyth (1984). This discrepancy may be due to the lower dose of PHA and older age of the birds in this study, as well as the genotypes of birds tested. Similar to the present study, Van der Zijpp (1983), Lamont and Smyth (1984), as well as Goto et al (1978) used conventionally housed birds for the PHA inoculation experiments.
The correlations of MD incidence with PHA response (table V) were generally low and inconsistent. The two correlations that were significantly different from zero, as well as the sign of 7 out of the 8 correlations for sexes combined in the table, suggested a tendency for the high PHA response to be associated with MD resistance. Consistent with the observations of Thompson et al (1980), this would indicate that the bird's ability to mount a cell-mediated immune response is important for MD resistance.
Of the three sets of Rand P-strains, described, for combined sexes, the Rstrains 2R and 3R had greater PHA response than the corresponding P-strains 2 and 1 (table II). The reverse was observed for the third set of R-strains 8R and P-strains 8. In addition most of the correlations between PHA response and MD resistance were non significant (table V). Therefore, in agreement with Calnek et al (1989), who studied in vitro PHA response of lymphocytes from strains of varying resistance to MD, we conclude that swelling response to PHA inoculation in the wing web is not sufficiently predictive of MD resistance to justify its use in genetic selection.