|
|
||||||||

* Department of Animal and Nutritional Sciences, University of New Hampshire, Durham, New Hampshire 03824; and
Department of Biological Sciences, Northern Illinois University, DeKalb, Illinois 60115
| Abstract |
|---|
|
|
|---|
Key Words: erythrocyte alloantigen major histocompatibility (B) complex oncogene Rous sarcoma virus tumor
| Introduction |
|---|
|
|
|---|
Gilmour (6) and Briles (7) independently discovered the L alloantigen system and established that two haplotypes, L1 and L2, segregate. The L locus segregates independently from nine other erythrocyte alloantigen systems (1), but has not been assigned to a linkage group. Alloantigen typing showed that the L system did not segregate (8) in a reference population established for molecular mapping. In addition to the described L haplotype associations with responses to Rous sarcomas virus (RSV)-induced tumors, other investigations have revealed L allele frequency alterations following divergent selection for bursa size (9) and fertility changes among different L genotypes (10). Taylor and Briles (11) examined the effect of eight non-MHC alloantigen systems on resistance and susceptibility to Eimeria tenella in both B2B2 and B2B5 backgrounds. The authors found an association only between the L system and cecal lesions. B2B2L1L1 chickens had higher lesion scores than B2B2L1L2 and B2B2L2L2 chickens. No significant L genotype effect was observed in a B2B5 background.
Rous sarcoma is a connective tissue tumor caused by the RSV, an oncogenic RNA virus. Tumors develop after injection of the virus into susceptible chickens. The tumors may regress or progress depending on the level of antitumor immune response produced by the MHC (12, 13). Variation in RSV tumor outcome among identical MHC genotypes from crosses of inbred lines differing at the MHC (12, 1417), among different inbred lines identical at the B complex (1820), or among crosses of noninbred lines (21, 22) implicated a role for non-MHC genes. For example, non-MHC T lymphocyte alloantigens, Ly-4 and Th-1, and the B lymphocyte alloantigen Bu-1 interacted to alter the response against RSV tumors in crosses of B2B2 inbred lines (19, 20).
Two previous studies found L alloantigen effects on the response to Rous sarcomas. Collins (23) studied the effect of alloantigen systems C, D, E, I, and L on tumor outcome in the F2 generation of inbred lines 63 x 100 that all had the B2B2 genotype. The C, D, E, and I systems did not influence tumor outcome. On the other hand, the L genotype significantly affected tumor growth in females, but not in males. LePage et al. (24) examined the effect of alloantigen systems A, C, D, E, H, I, L, and P on Rous sarcoma outcome in two B complex genotypes: B5B5, a tumor progressor, and B2B5, a moderate progressor. Alloantigen systems A, C, D, E, H, I, and P had no significant effect on tumor fate. The L genotype correlated with a differential tumor outcome. In the B2B5 genotypic background, the L1L1 chickens had lower tumor size, TPI, and mortality than the L1L2 chickens. Mortality was lower in L1L1 birds compared with L1L2 birds in the B5B5 background.
These earlier experiments indicate significant L alloantigen modulation of immune responses without complete B and L system segregation. The objective of this study is to further investigate L system effects on RSV-induced tumors. We used crosses producing progeny fully segregating for both the B complex and the L alloantigen. This structure allows examination of the L system effects in B genotypes that vary widely in their RSV-induced tumor outcome.
| Materials and Methods |
|---|
|
|
|---|
Line 6.15-5 (B5B5 L1L1) dams were mated to B2B2 L1L2 sires from a line cross between modified Wisconsin Line 3 Ancona x White Leghorn line NIU 4 sires (B2B2 L1L2), as described by LePage et al. (24). Chickens from this mating that had the B2B5 L1L2 genotype contained 50% of the Line 6.15-5 genome and were used as parents to produce the experimental progeny. Pedigree matings of two B2B5 L1L2 sires to five B2B5 L1L2 dams per sire produced six hatches having one hundred fifty-one chicks segregating for all possible combinations of B and L genotypes. The birds were hatched at the University of New Hampshire Poultry Research Farm and were wing-banded for identification. Vaccinations against Marek's disease and Newcastle bronchitis were administered at hatch and 10 days, respectively. The chicks were housed in heated brooder batteries with water and food freely available. Six-week-old chicks were transferred to isolation cages for the remainder of the experiment.
Alloantigen Typing.
The chickens were typed for B and L systems in agglutination assays utilizing antisera specific for the haplotypes of the parental stocks (27). When chicks reached 3 weeks of age, 0.5 ml of blood was drawn from the wing vein to cold sodium citrate anticoagulant solution (68 µM sodium citrate/72 µM sodium chloride). Samples were shipped overnight with ice packs to Northern Illinois University. Fifty microliters of a 2% suspension of washed red blood cells was dispensed into tubes containing 100µl of antiserum specific for the B and L system haplotypes of interest. Following a 2-hr room-temperature incubation, the reaction mixtures were transferred to 3°C for an overnight incubation. The following day, cells were resuspended and scored visually for agglutination after a 1-hr incubation at room temperature.
RSV Challenge and Tumor Evaluation.
At six weeks of age, the birds were inoculated in the right wingweb with 20 pfu of the Bryan high-titer strain of RSV (RAV-1), subgroup A. Two weeks following RSV challenge, tumors were scored for size using the following scale: 0, no palpable tumor; 1, small tumor up to 0.5 cm in diameter; 2, tumor >0.5 up to 1.2 cm in diameter; 3, tumor >1.2 cm up to one-half of wingweb area; 4, tumor > one-half of wingweb area, but < entire wingweb area; 5, tumor filling the entire wingweb; 6, massive tumor extended beyond wingweb; and 7, death during the experiment (12). Tumor size was also scored at weeks 3, 4, 6, 8, and 10 postinoculation for a total of six tumor size scores over the 10-week experimental period. The six tumor size scores were then used to assign a TPI to each bird as an indicator of the tumor growth pattern. The TPI values were those of Collins et al. (28) where 1 = complete regression by 70 days postinoculation, or a decreasing slope, or complete regression by 56 days followed by recurrence; 2 = general upward trend, or plateau; slight regression after 56 days; 3 = terminal tumor after 42 days postinoculation; 4 = terminal tumor between 29 and 42 days postinoculation; and 5 = terminal tumor by 28 days postinoculation.
Statistical Analysis.
Tumor scores were analyzed by repeated measures analysis of variance (ANOVA) with hatch, sex, sire, dams within sire, time, B genotype, L genotype, and a B x L interaction as main effects. A large effect of B type led to separate analyses for each B genotype having hatch, sire, dams within sire, time, and L type as main effects. The TPI values were rank transformed and analyzed by ANOVA as described by Conover and Iman (29) with the same independent variables except time as in the repeated measures ANOVA. Significant differences between alloantigen system genotypes were determined using Fisher's protected LSD. Mortality rates were evaluated using chi-square analysis.
| Results |
|---|
|
|
|---|
|
|
The L genotype, however, exerted a significant influence on tumor score, TPI, and mortality in the B5B5 genotypic background. Repeated measures analysis of tumor score indicated a significant change in tumor size over time (P = 0.0001), as well as a significant L genotype x time interaction (P = 0.00017). The overall pattern of tumor growth was an increase in tumor size 2, 3, and 4 weeks postinoculation, followed by a lower rate of tumor size increase in B5B5 L1L2 (n = 14) compared with either the B5B5 L1L1 (n = 5) or B5B5 L2L2 (n = 12) genotypes (Fig. 2A
). Tumor size at 10 weeks postinoculation was diminished in the L1L2 genotype (5.43 ± 0.6) compared with either the L1L1 (7.00 ± 0.0) or L2L2 (6.83 ± 0.2) genotypes.
|
| Discussion |
|---|
|
|
|---|
The L alloantigen genotypes were analyzed within each B genotype to eliminate the possibility that the powerful B complex effect would overcome any L system effects. No significant L effect was evident in either the B2B2 or B2B5 genotypic background, as both genotypes regressed tumors. A significant L influence, however, was evident in the B5B5 background. This result suggests that the robust tumor regression found in B2B2 and B2B5 birds masks the weak L effect, which becomes obvious only in the B5B5 progressor background. Based upon tumor growth over time, TPI, and mortality, the L1L2 genotype mitigated the progressive effect of the B5B5 genotype and did so significantly compared with the L1L1 genotype.
Two prior studies examined the L alloantigen system influence on Rous sarcomas. Among the F2 generation of B2B2 inbred lines 63 x 100, tumor fate was affected significantly by L genotype (23). Females of the L1L1 genotype had significantly lower TPI than the L2L2 genotype. No L genotype effect was found in males. LePage et al. (24) tested B2B5 and B5B5 progeny that segregated for at least two alleles of eight non-B alloantigen systems. Significant L alloantigen effects on Rous sarcomas were found in both B complex backgrounds. The L1L1 genotype was associated with lower tumor scores, TPI, and mortality than the L1L2 genotype in a B2B5 background. Lower mortality was found in B5B5 L1L1 chickens than in B5B5 L1L2 chickens.
The apparent disagreement of the present data with previous studies may be attributed to several factors. First, each study used a mating type with a different genetic background. Collins (23) used progeny from the cross of two inbred lines, LePage et al. (24) used progeny that were 50% Modified Wisconsin Line 3 and 50% White Leghorns, and the current study used progeny that consisted of 50% inbred line 6-15.5 and 50% Modified Wisconsin Line 3 x NIU 4 White Leghorns. Second, LePage et al. (24) used a higher virus dose (30 pfu) than this experiment (20 pfu). Virus dose may influence tumor growth in progressive genotypes (37), such as B5B5. Third, the current study is the only one that utilizes full segregation of B and L haplotypes. Effects of background genes, other than B and L, cannot be completely excluded by the results.
Growth and subsequent division of tumor cells as well as cellular recruitment via viral-replication genes (30) affect RSV tumor growth. T cells are principally responsible for RSV tumor regression (17, 19, 38, 39). Cross-reactions between tumor antigens and certain MHC haplotypes may impinge on antigen recognition. The B5 antigen cross-reacted to one or more RSV tumor antigens because Rous sarcoma progression increased in B2B2 chickens previously made tolerant to the B5 antigen from progressor chickens compared with untreated B2B2 controls (40, 41). Tumor regressor Line CB B12B12 chickens progressed a transplantable v-src-induced tumor after they were made tolerant to Line CC B4B4 or CB.R1, (B-F12 B-G4). This result supported a cross-reaction between the B4 antigen and a v-src tumor antigen (36).
The L1L2 genotype advantage in B5B5 tumor outcome indicates complementation. The host response to a tumor is a complex reaction to a multitude of antigens (42). Particular heterozygous combinations of MHC molecules may facilitate tumor or viral antigen recognition (43). Compared with the homozygotes, a heterozygote may complement through either improved recognition efficiency of the same number of antigens or increased recognition of additional antigens. The L1L2 genotype or closely linked genes, in B5B5 chickens of the current study, may have enhanced the immune response against RSV tumors by interacting with an effector molecule, facilitating viral or tumor antigen interaction with effector molecules or partially overcoming the B5 antigen-associated tolerance to tumor antigens. These effects might be accomplished through increased T cell activation, B complex antigen expression, or both.
Plachy (44) described MHC and non-MHC gene complementation in crosses of the Prague congenic inbred lines, CB (B12B12), CB.I (B7B7), and inbred line IA (B7B7). Chickens of the B12B7 genotype from either CB x CB.I or CB x IA matings had more regressing tumors than either B12B12 or B7B7 chickens, indicating complementation between the B12 and B7 haplotypes. In addition, greater tumor regression was evident in the B12B7 genotype CB x IA chickens than in the same genotype from the CB x CB.I mating, denoting a non-MHC gene effect that was complementary. Matings of Line CC (B4B4) with CB.I and IA revealed similar MHC and non-MHC complementation that was independent of age. B4B7 chickens from the CC x IA mating had more tumor regression than CC x CB.I cross (44).
Other non-MHC genes have demonstrated complementary effects on the outcome of Rous sarcomas. Gilmour et al. (19) studied progeny derived from inbred line crosses identical for the MHC (B2B2) but segregating for Ly-4 and Th-1 non-MHC T cell-surface antigen genes. A homozygous/heterozygous interaction was found in that Ly-4aLy-4a/Th-1aTh-1b and Ly-4aLy-4b/Th-1aTh-1a genotypes had increased RSV tumor regression. The interaction between the a and b haplotypes occurred when the other locus was the aa genotype. Another example of complementation occurred between non-MHC B and T cell alloantigens, Bu-1 and Ly-4. Progeny from a different cross of B2B2 inbred lines had greater RSV tumor regression due to complementation between the Ly-4a and Bu-1b alleles or the Ly-4b and Bu-1a alleles (20).
Genes other than the MHC can affect the RSV tumor progressive B5B5 genotype. Collins et al. (22) examined Rous sarcoma metastasis in B5B5 hosts from two populations: (Line 61 x Line 151) F5 White Leghorn cross and (Line 61 x Line 151) F5 Leghorn x Line UNH 105 New Hampshire F2. The incidence of tumor metastasis was significantly lower in the B5B5 White Leghorn x UNH 105 cross than in the B5B5 White Leghorn population, suggesting a possible non-MHC background effect on metastasis. Another study found that B5B5 birds having alloantigen haplotypes D1+ or I8+ had significantly lower TPI than those with D1- or I8- (24). Mortality in B5B5 chickens was lowered by the L1L1 genotype compared with the L1L2 genotype (24). However, that study used chickens with a different genetic background and used two B genotypes with three L genotypes compared with the current research that has full segregation of B and L genotypes.
The present experiments revealed that the B complex and the L alloantigen system significantly affected tumor growth over time, TPI, and mortality due to RSV-induced tumors. This work also adds unique information regarding L alloantigen effects on the immune response to Rous sarcomas in the context of B2 and B5 haplotype segregation. Complementation by the L1L2 genotype in the B5B5 birds supports the conclusion that the L alloantigen system, or some closely linked gene(s), has significant effects on the fate of Rous sarcomas. Further research should examine possible interactions between the L system, different B haplotypes, and other non-MHC genes.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 To whom requests for reprints should be addressed at Department of Animal and Nutritional Sciences, University of New Hampshire, Kendall Hall, Durham, NH 03824. E-mail: bob.taylor{at}unh.edu ![]()
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |