Major Histocompatibility Complex Class IV Restriction Fragment Length Polymorphism Markers in Replicated Meat-Type Chicken Lines Divergently Selected for High or Low Early Immune Response

Major Histocompatibility Complex Class IV Restriction Fragment Length Polymorphism Markers in Replicated Meat-Type Chicken Lines Divergently Selected for High or Low Early Immune Response ZEHAVA UNI,1-2 MICHAL GUTMAN,3 GABRIAL LEITNER,1 ESTHER LANDESMAN,1 DAN HELLER,1 and AVIGDOR CAHANER3 Department of Animal Science and Department of Genetics, Faculty of Agriculture, The Hebrew University of Jerusalem, Rehovot 76100, Israel ABSTRACT Information on MHC may improve the efficiency of selection for immunological traits via the application of marker assisted selection or by selecting directly for a specific restriction fragment length polymorphism (RFLP) band or MHC haplotype. An experimental procedure is presented here for identifying MHC genes that are related to early immune response. A Class IV cDNA clone was used to probe Southern blots of erythrocyte genomic DNA from chickens. Chickens were taken from the second (S2) and third (S3) generations of replicated lines divergently selected for high antibody response (HC1, HC2) or low antibody response (LCI, LC2) to Escherichia coli vaccination at 10 days of age. These selection criteria have been found to be associated with other immunological parameters. The hypothesis that these selected lines differ in their MHC loci was evaluated by comparing the frequencies of MHC RFLP markers (single RFLP bands) and haplotypes (patterns of RFLP bands). The significant differences between LC and HC in the frequency of many MHC RFLP bands and of five MHC haplotypes indicate that early antibody production is influenced by MHC genes. The reliability of the association between the selection and frequency differences was tested and proven in most cases by analysis of the replicated lines. These differences in RFLP markers represent a change in allelic frequencies in MHC genes, probably due to selection. The results imply a connection between the Class IV genes and early antibody production, and they show the potential of prospective breeding not only by immunological phenotype but also by genotype (i.e., using RFLP markers of the MHC). (Key words: chicken, restriction fragment length polymorphism, major histocompatibility complex, Class IV, immune response) 1993 Poultry Science 72:1823-1831

INTRODUCTION The incidence of disease in commercial chicken flocks is of major economic concern. Vaccines, antibiotics, and other drugs are being used in the poultry industry in

Received for publication January 4, 1993. Accepted for publication June 10, 1993. iDepartment of Animal Science. 2 To whom correspondence should be addressed. 3 Department of Genetics.

order to avoid economic loss. It is clear that immunocompetence and resistance to many infectious diseases is in part genetically determined (Gavora and Spencer, 1983). Consequently, there has been increasing interest, in the identification of genetic and immunological markers that could be used in breeding programs to improve immunocompetence. The usefulness and limitations of marker-assisted selection (MAS) have been discussed by Soller and Backmann

1823

1824

UNI ET AL.

(1986) and Smith and Simpson (1986). The more markers identified and the closer they are to quantitative trait loci, the more efficient MAS is likely to be. When a certain marker or group of markers is associated with a quantitative trait that is important to the breeder, MAS is more advantageous than conventional selection. Developments in molecular biology techniques have enabled more efficient evaluation of potential markers in the genome. In some species of farm animals, strong evidence has been found for the linkage between restriction fragment length polymorphism (RFLP) and prolactin (Cowan et al, 1989), production traits (Jung et al, 1989), and disease resistance (Lunden et al, 1990). However, in chickens, polymorphisms identified by the use of DNA technology and associations between RFLP markers and quantitative traits have hardly been described. High polymorphism of RFLP markers was revealed in the MHC Class IV (B-G) region following digestion of DNA from layer chickens (Miller et al, 1988) and meat-type chickens (Uni et al, 1992). These RFLP are potential markers for MHC genes. The MHC genes play an important role in immune system functions (reviewed by Bacon, 1987). In chickens, the MHC (B complex) is located on an 8 million bp microchromosome, of which 6 million bp code for ribosomal RNA in the nuclear organizer (Bloom and Bacon, 1985). The MHC contains tightly linked genes B-L, BV, and B-G (Hala et al, 1988; Kaufman et al, 1989) that code, respectively, for Class I, Class II, and Class IV glycoproteins on the cell surface. The B-F and B-L antigens are similar to their mammalian homologues in structure, function, and tissue distribution (Guillemont and Auffray, 1989). The B-G antigen, however, is unique to avian species and was originally described as limited to erythrocytes and their precursor cells (Pink et al, 1977). Recent publications, however (Miller et al, 1990; Salomonsen et al, 1991a) have shown that members of this superfamily of immunoglobulins are also found on thrombocytes, lymphocytes, and certain epithelial cells in the bursa of Fabricius, thymus, and intestine. The grouping of these genes on one chromosome is termed a B-haplotype.

Different B-G haplotypes have been numbered from Ba to Bn (Briles and Briles, 1982) and used for genetic studies. Most of those studies have been conducted in White Leghorn (WL) populations, using serological tools for B-G haplotype identification. Several studies have demonstrated the association between the MHC and oncogenic diseases (Bacon et al, 1981) or bacterial disease challenge survival (Lamont et al, 1987). Nordskog (1983) and Pevzner et al. (1975) reported different levels of antibodies in different B haplotypes. Based on all of the above, the MHC region is a natural candidate for MAS, especially for immunological traits. In the present study, cDNA clone bg 32.1 (Miller et al, 1988) was used to probe Southern blots of erythrocyte genomic DNA from chickens. The examined chickens were taken from the second (S2) and third (S3) generations of replicated lines divergently selected for high antibody response (HC1, HC2) or low antibody response (LCI, LC2) to Escherichia coli vaccination at 10 days of age (Leitner et al, 1992). These selection criteria have been found to be associated with other immunological parameters, such as antibody response to Newcastle disease virus (NDV) inactivated vaccine, SRBC, increased phagocytic activity, and proliferative response to antigens or mitogens (Pitcovski et al, 1987; Heller et al, 1992). The selection therefore most probably affects a general component of early immiinocompetence, possibly by changing allelic frequencies in MHC genes. The hypothesis that these selected lines differ in their MHC loci was evaluated by comparing the frequencies of MHC RFLP markers (single RFLP bands) and haplotypes (patterns of RFLP bands). MATERIALS AND METHODS Birds

The analysis included 102 chickens from the S2 generation and 102 chickens from the S3 generation of Lines HC1, HC2, LCI, or LC2. The immunological parameters of these selection lines are presented in Heller et al. (1992).

CLASS IV MARKERS AND CHICKEN IMMUNE RESPONSE

Restriction Fragment Length Polymorphism of Class IV Major Histocompatibility Complex

1825

verify the consistency of the selection effect in those replicated lines.

The DNA was extracted from samples of heparinized whole blood collected from each chicken (Hillel et d., 1989). Genomic DNA (10 /xg) was digested with the restriction endonucleases Pvull (83 chickens from S2 generation and 100 chickens from S3 generation) or Bglll (102 chickens from S2 generation and 100 chickens from S3 generation) prior to electrophoresis on a 1% agarose gel. The DNA was transferred to nylon filters according to the method of Southern (1975), hybridized (Hillel et ah, 1989) with the 32P-labeled chicken Class IV cDNA probe bg 32.1 (Miller et al., 1988), and subjected to autoradiography.

RESULTS Frequencies of Restriction Fragment Length Polymorphism of Class IV Major Histocompatibility Complex

Digestion of DNA with Restriction

Enzyme Pvull. An RFLP analysis following digestion with the restriction enzyme Pvull revealed high levels of polymorphism and complexity at the Class IV region. A total of 31 polymorphic bands (1.6 to 8.0 kb) were observed in 183 chickens, taken at random from S2 and S3 generation birds of lines LCI, LC2, HC1, and HC2. Twelve of these RFLP bands were analyzed statistically. In S2, five RFLP bands (P2, P4, P6, P24, Statistical Analysis P25) had significantly different frequencies Differences of band and haplotype fre- in the HC versus LC lines (Table 1). In S3, six quencies between the pooled, divergently RFLP bands (P3, P4, P22, P23, P24, P25) selected lines (LC and HC) were analyzed were found to have significantly different by chi-square test using PROC FREQUEN- frequencies in the divergent lines (Table 2). CIES (SAS Institute, 1987). In those cases in For Band P22, the significant difference which LC differed significantly from HC, between HC1 and HC2, and the similarity the differences between replicated lines between HC2 and LCI, suggests that the were also analyzed by chi-square test to difference between pooled LC and HC was

TABLE 1. Frequency of 12 RFLP1 bands obtained using restriction enzyme PPWII in the second generation of chicken lines selected for high (HC) or low (LC) antibody response to Escherichia coli Replicated lines RFLP band

2

LCI

LC2

HC1

(30)3

(30)

(48)

HC2

LC

HC

(58)

(60)

(106)

20 23 0 16 97 41 23 27 27 13 22 15

13 9 6 2 94 27 24 17 23 10 7 2

P(X2)

C°'1

(kb) 8.0 7.5 5.2 4.7 4.5 4.3 2.3 2.0 1.9 1.8 1.7 1.6

Pooled lines

PI P2 P3 P4 P5 P6 P20 P21 P22 P23 P24 P25

23 26 0 16 100 47 33 20 18 16 17 13

16 20 0 20 94 37 13 33 13 10 27 16

14 4 8 2 100 33 23 12 23 16 12 0

Restriction fragment length polymorphism. 2P1... P25 designate the different RFLP bands. 32n.

(—)

12 12 3 4 90 28 24 20 24 5 4 7

NS .004 .056 <.0001 NS .020 NS .085 .067 NS .004 .001

1826

UNI ET AL. TABLE 2. Frequency of 12 RFLP1 bands obtained using restriction enzyme PPMII in the third generation of chicken lines selected for high (HC) or low (LC) antibody response to Escherichia coli Pooled lines

Replicated lines RFLP band 2

LCI

LC2

HCl

(52)3

(58)

(32)

19 26 4 15 100 37 27 30 19 2 21* 13

14 12 2 12 82 44 7 37 10 7 36 7

22 6 9 0 100 25 25 50 37* 13 9* 0

(kb) 8.0 7.5 5.2 4.7 4.5 4.3 2.3 2.0 1.9 1.8 1.7 1.6

PI P2 P3 P4 P5 P6 P20 P21 P22 P23 P24 P25

HC2 (58) (°/,^ 13 20 22 5 86 38 33 36 19 19 0 0

LC

HC

(110)

(90)

16 14 3 14 89 40 17 34 14 4 29 20

17 16 18 4 91 33 33 41 25 17 3 0

P(x2)

NS NS <.0001 .008 NS .082 NS NS .025 .003 <.0001 .001

'Restriction fragment length polymorphism. P 1 . . . P25 designate the different RFLP bands. 32n. 'Indicates a significant difference (P < .05) between a pair of replicated lines. 2

random and not due to selection. Although the replicated lines differed significantly in the frequency of Band F24, its frequency in LCI and LC2 was much higher than that in HCl and HC2 (Table 2). Digestion of DNA with Restriction Enzyme BgMI. The use of a second sixcutter restriction enzyme revealed a similar degree of polymorphism and complexity at the Class IV region. A total of 25 polymorphic bands (2.0 to 9.5 kb) were observed in an experimental population of 202 chickens taken at random from the S2 and S3 generations of the replicated divergent lines. Seven of the 25 RFLP bands were analyzed statistically. In S2, four RFLP bands (B19, B20, B21, and B24) had significantly different frequencies in LC and HC birds (Table 3). For Band B19, the significant difference between HCl and HC2, and the similarity between HCl, LCI, and LC2 suggest that the difference between pooled LC and HC was random and not due to selection. Although the HC replicated lines differed significantly in Band B20, its frequency in LCI and LC2 was much higher than that in HCl and HC2. Statistical analysis of pairs of replicated lines showed that the difference between the pooled LC

and HC in RFLP Band B20 could be random and not due to selection. In S3 generation birds, the frequency of these RFLP bands and of another band (B25) were found to be significantly different in the divergent lines (Table 4). Frequencies of Major Histocompatibility Complex Class IV Restriction Fragment Length Polymorphism Haplotypes

Haplotypes were determined according to their "restriction pattern" and by family group analysis, as described by Uni et al. (1992) and Landesman et al. (1993). A total of 14 different haplotypes (A, C, D, E, F, G, H, K, L, R, T, U, V, and X) were identified and designated by letters in 102 chickens from the replicated divergent lines (LCI, LC2; HCl, HC2) of S3 generation birds. Among the chickens examined, a large variety of B-G haplotypes in all kinds of combinations was found. Only 14% of the chickens were homozygous. Figure 1 exhibits a representative Southern blot depicting the RFLP pattern of 19 chickens from two selection lines (LCI; HCl), after digestion with Pvull (Figure la) or BglU (Figure lb) and hybridization with Class IV probe

CLASS IV MARKERS AND CHICKEN IMMUNE RESPONSE

1827

1

TABLE 3. Frequency of seven RFLP bands obtained using restriction enzyme BglU in the second generation of chicken lines selected for high (HC) or low (LC) antibody response to Escherichia coli Replicated lines RFLP band*

LCI

LC2

HC1

(52)3

(52)

(42)

HC2

LC

HC

(58)

(104)

(100)

33 24 21 39 4 7 25

30 14 4 46 5 32 25

P(x2)

(°',\

(kb) 2.9 2.8 2.5 2.4 2.2 2.1 2.0

Pooled lines

B18 B19 B20 B21 B23 B24 B25

40 24 25 42 0 6 26

24 24 17 35 7 7 24

43 24* 8* 48 4 30 31

21 7 0 44 5 35 20

NS .037 .004 .012 NS <.O01 NS

Restriction fragment length polymorphism. B18...B25 designate the different RFLP bands. 32n. 'Indicates a significant difference (P < .05) between a pair of replicated lines. 2

bg 32.1. This figure also demonstrates the designation of the B-G haplotypes and some of the restriction fragment bands that were analyzed statistically. Among the seven chickens from the LC line, five chickens were carrying Haplotype K in different combinations (with Haplotype R, A, or G). Among the 12 chickens from the HC line, six were carrying Haplotype D in different combinations (with Haplotype V or R). Of these 19 chickens, 5 were homozygous. The interpretation of these RFLP patterns as

haplotypes of the MHC region were followed by evaluation of their frequencies within the selection lines. Only haplotypes that were found in at least 8 birds out of the 102 examined chickens were analyzed statistically. Several haplotypes (A, C, F, K, U, and X) appeared only or mainly among either LC or HC birds (Table 5). With the exception of Haplotype K (statistical analysis between pairs of replicated lines showed that the difference between the pooled LC and HC could be random and not due to the

TABLE 4. Frequency of seven RFLP1 bands obtained using restriction enzyme BglU in the third generation of chicken lines selected for high (HC) or low (LC) antibody response to Escherichia coli Pooled lines

Replicated lines RFLP band* (kb) 2.9 2.8 2.5 2.4 2.2 2.1 2.0

LCI

LC2

HC1

(54)3

(58)

(32)

HC2

LC

HC

(56)

(112)

(88)

44 7 0 50 3 38 16

34 23 13 29 6 8 26

42 8 1 47 4 38 16

P(x2)

(o/\

B18 B19 B20 B21 B23 B24 B25

43 22 7 33 6 9 22

25 30 18 26 6 7 31

38 11 4 37 6 40 15

Restriction fragment length polymorphism. 2 B18 . . . B25 designate the different RFLP bands. 32n.

.074 <.001 <.001 <.001 NS <.O01 .046

1828

UNI ET AL. HC

LC Kt

10

4 5

1

11 12 13 14 15

16

17 18 19

5.14.23.5-

2-

Ml U p . , KR

AK

1

AC KK KG KG DO DV DV XV CC DD GX UG VX TU DD DR LC

1

2

3

4

HC 5

6

V

•8

9

10

11

12

13

14

15

16

17

18

19

H

!liifflt*S|

6 4-

• .

B ir • - W 820

2.3-

W

KR

•

^ ^ 5

• - i — #

AK

• B21

" f I - £

•B24

#

•

w

•

w

AC KK KG KG DD DV DV XV CC DD GX UG VX TU OD DR

FIGURE 1. A representative Southern blot depicting the restriction fragment length polymorphism patterns obtained following digestion with PvuU (a) and BglTl (b) of genomic DNA (from chicken lines selected to high (HC) or low (LC) antibody response to Escherichia coli (LC = 7 and HC = 12 chickens) and hybridization with Class IV probe bg 32.1. Restriction fragment sizes were estimated on the basis of Hmdm-digested 5-phage DNA fragments. Some RFLP fragments are marked. P2, P3, P17, P20, P24, P25 on Figure la and B19, B20, B21, B24, B25 on Figure lb are Class rv RFLP fragment numbers (P for PuuII and B for Bg/II). Class IV Haplotype designation is indicated at the bottom of each lane.

1829

CLASS IV MARKERS AND CHICKEN IMMUNE RESPONSE

TABLE 5. Frequency of B-G haplotypes in the chicken lines selected for high (HC) or low (LC) antibody response to Escherichia coli Pooled lines

Replicated lines Haplotypes

LCI

LC2

HCl

(54)1

(58)

(34)

HC2

LC

HC

(58)

(112)

(92)

2 0 28 8 14 18 0 4 14 12 3.8

11 13 24 15 3 8 21 4 0 0 1.2

1 0 20 12 12 17 4 11 11 9 3.6

P(x2)

u

A C D E F G K R U X Ln titer 2

16 12 30 16 6 8 12* 0 0 0 .9

7 15 18 15 0 9 30 7 0 0 1.3

0 0 6 20 10 16 10* 23 6 6 3.5

\' )

.005 <.001 NS NS .025 NS <.001 NS <.001 .005

i2n. 2

Line means of antibody response to E. coli vaccination at 10 days of age. 'Indicates a significant difference (P < .05) between a pair of replicated lines.

selection), all those haplotypes were significantly different in their frequency in the selection lines. DISCUSSION The field of molecular biology has provided new tools and approaches to animal breeding, including identification of genes influencing quantitative traits. An important condition for finding a marker for a given trait is to use a probe that reveals a high level of polymorphism in the gene region examined. The bg 32.1 probe, which binds to the B-G genes, reveals a very high degree of polymorphism (Miller et al, 1988; Uni et al, 1992) and is therefore a suitable marker. Based on serological and biochemical information, B-G genes appear to exist within a subregion that is physically separated from B-F/B-L genes (Briles and Briles, 1982; Miller et al, 1988). However, Kaufman et al (1989) raised the question of whether all B-G genes are separated from all B-F/B-L genes. His evidence indicates that some B-G genes are located within the B-F/B-L subregion. His model introduced the possibility that RFLP fragments include Class I or Class II and Class IV genes and can therefore be identified by a Class IV probe. '

Miller et al. (1991) recently explained the nature of the high degree of polymorphism in B-G antigens. They found that the B-G polypeptides are composed of single extracellular domains and single membrane-spanning domains (all of them with sequence diversity) and long cytoplasmic tails, which are made up entirely of differing numbers of units, each consisting of seven amino acid residues (heptads), which may themselves vary in sequence. The large number of RFLP bands obtained in the current study agrees with their findings. Moreover, A PvuU restriction site was located in the sequence of multiple short (42 bp) repeats that had been reported by Kaufman et al. (1989). These repeats have been suggested to have a function in the size heterogeneity of B-G molecules, due to variation of their cytoplasmic tail length. The most polymorphic genes known in eukaryotic organisms are associated with the immune system functions of antigen presentation and recognition. The findings of Miller et al. (1991) revealed that the B-G genes are similar to immunoglobulin, having extracellular domains with variable sequences. Examination of the immune function of B-G antigens (Salomonsen et al, 1991b) showed an "adjuvant effect" of chicken MHC polymorphic B-G antigens and suggested rules for the polymorphic

1830

UNI ET AL.

B-G molecules, particularly for the generation of B cell diversity, in the immune systems of birds. The recent information about the function of B-G antigens emphasizes the potential usefulness of B-G RFLP markers in selection for improved immunocompetency. The significant differences between LC and HC in the frequency of many MHC RFLP bands and of five MHC haplotypes indicate that early antibody production is influenced by MHC genes. The reliability of the association between the selection and frequency differences was tested and proven in most cases by analysis of the replicated lines. These differences in RFLP markers represent a change in allelic frequencies in MHC genes, probably due to tie selection. The increase in frequency differences between LC and HC from S2 to S3 observed for most RFLP bands provides an additional indication of the association between the selection and the change in band frequencies. Information on MHC may improve the efficiency of selection for immunological traits, via the application of MAS or by selecting directly for a specific RFLP band or MHC haplotype. The use of two restriction enzymes increased the number of RFLP markers, improved the definition of MHC haplotypes (unpublished data), and increased the chances of identifying an association with immune traits. If a certain fragment or group of fragments (i.e., a haplotype) were to be associated with a trait and if its effect were to be large enough, MAS would be more advantageous than conventional selection. Analyses of the duplicated selection lines, which differ significantly in several immunological parameters at young age (Heller et ah, 1992), enabled us to compare the frequencies of RFLP markers. For the first time in meat-type chickens, a significant association between frequency of Class r / MHC RFLP markers and immune response was found. Differences between LC and HC lines suggest some linkage between the RFLP markers and the immune characteristics of the lines. The results imply a connection between the Class IV genes and early antibody production, and they show the potential of prospective breeding not only by im-

munological phenotype but also by genotype (i.e., using RFLP markers of the MHC). ACKNOWLEDGMENTS

The authors thank M. M. Miller (Beckman Research Institute, Duarte, CA 91010) for the bg 32.1 probe and J. Pitcovski (Megal Technology Center, Kiriat-Shmona, Israel) for his contribution. This work was funded by Bundesministerium fur Forschung und Technologie, Germany; Ministry of Science and Technology, Israel; and Joint German Israel Research Projects. REFERENCES Bacon, L. D., 1987. Influence of the major histocompatibility complex on disease resistance and productivity. Poultry Sci. 66:802-811. Bacon, L. D., R. L. Witter, L. B. Crittenden, A. Fadly, and J. Motta, 1981. B-haplotype influence on Marek's disease, Rous sarcoma and lymphoid leukosis virus-induced tumors in chickens. Poultry Sci. 60:132-139. Bloom, S. E., and L. D. Bacon, 1985. Linkage of the major histocompatibility complex and the nucleolar organizer in the chicken. J. Hered. 76: 146-154. Briles, W. E., and R. W. Briles, 1982. Identification of haplotypes of the chicken major histocompatibility complex (B). Immunogenetics 15:449-459. Cowan, C. M., M R. Dentine, R. L. Ax, and L. A. Schuler, 1989. Restricted fragment length polymorphism associated with growth hormone and prolactin genes in Holstein bulls: evidence for a novel growth hormone allele. Anim. Genet. 20: 157-165. Gavora, J. S., and J. L. Spencer, 1983. Breeding for genetic resistance: specific or general. World's Poult. Sci. J. 43:137-148. Guillemont, F., and C. Auffray, 1989. Molecular biology of the chicken major histocompatibility complex. Crit. Rev. Poult. Biol. 2:255-275. Hala, K., A.-M. Chausse, O. Bourlet, O. Lassila, V. Hasler, and C. Auffray, 1988. Attempt to detect recombination between B-F and B-L genes within the chicken B complex by serological typing, in vitro MLR and RFLP analyses. Immunogenetics 28:433-438. Heller, E. D., G. Leitner, A. Friedman, Z. Uni, M. Gutman, and A. Cahaner, 1992. Immunological characteristics of meat-type chicken lines divergently selected by antibody response to Escherichia coli vaccination at 10 days of age. Vet. Immunol. Immunopathol. 34:159-172. Hillel, J., Y. Plozky, A. Haberfeld, U. Lavi, A. Cahaner, and A. J. Jeffreys, 1989. DNA fingerprints of poultry. Anim. Genet. 20:25-35. Jung, Y. C, M. F. Rothschild, M. P. Flanagan, L. L. Christian, and C. M. Warner, 1989. Association of RFLP of swine leucocyte antigen class I genes

CLASS IV MARKERS AND CHICKEN IMMUNE RESPONSE

1831

with production traits of Duroc and Hampshire Nordskog, A. W., 1983. Immunogenetics as an aid to boars. Anim. Genet. 20:79-91. selection for disease resistance in the fowl. World's Poult. Sci. J. 39:199-209. Kaufman, J., J. Salomonsen, and K. Skjodt, 1989. B-G cDNA clones have multiple small repeats and Pevzner, I., A. W. Nordskog, and M. L. Kaeberle, 1975. Immune response and the B blood group hybridize to both chicken MHC regions. Imlocus in chickens. Genetics 80:753-759. munogenetics 30:440-451. Lamont, S. J., C. Bohn, and N. Cheville, 1987. Genetic Pink, J.R.L., W. Droege, K. Hala, V. C. Miggiano, and A. Ziegler, 1977. A three-locus model for the resistance to fowl cholera is linked to the major chicken major histocompatibility complex. Imhistocompatibility complex. Immunogenetics 25: munogenetics 5:203-216. 284-289. Landesman, E., Z. Uni, and E. D. Heller, 1993. Pitcovski, J., E. D. Heller, A. Cahaner, and B. A. Peleg, 1987. Selection for early responsiveness of Classification of Major Histocompatibility Comchickens to Escherichia coli and Newcastle displex class IV haplotypes by RFLP, in meat-type ease virus. Poultry Sci. 66:1276-1282. chickens. Anim. Genet, (in press). Leitner, G., Z. Uni, A. Cahaner, and E. D. Heller, Salomonsen, J., D. Dunon, K. Skjodt, D. Thorpe, O. Vainio, and J. Kaufman, 1991a. Chicken major 1992. Replicated divergent selection of meathistocompatibility complex-encoded B-G antitype chickens for high or low early antibody gens are found on many cell types that are response to E. coli vaccination. Poultry Sci. 71: important for the immune system. Proc. Natl. 27-37. Acad. Sci. USA 88:1359-1363. Lunden, A., S. Sigurdardottir, I. Edfors-Lilja, B. Danell, J. Redel, and L. Andersson, 1990. The Salomonsen, J., H. Eriksson, K. Skjodt, T. Lundgreen, M. Simonsen, and J. Kaufman, 1991b. The effect of bovine major histocompatibility com"adjuvant effect" of the polymorphic B-G antiplex on bull breeding values for clinical mastitis gens of the chicken major histocompatibility and somatic cell counts in milk. Pages 497-500 complex analyzed using purified molecules in: Proceedings of the 4th World Congress on incorporated in liposomes. Eur. J. Immunol. 21: Genetics Applied to Livestock Production. XVI, 649-658. Edinburgh, UK, July 23-27. SAS Institute, 1987. SAS/STAT® Guide for Personal Miller, M. M., H. Abplanalp, and R. Goto, 1988. Computers. 6th Edition. SAS Institute Inc., Cary, Genotyping chickens for the B-G subregion of NC. the major histocompatibility complex. Im- Smith, C, and S. P. Simpson, 1986. The use of genetic munogenetics 28:374-379. polymorphism in livestock improvement. J. Miller, M. M., R. Goto, S. Young, J. Chirivella, D. Anim. Breed. Gen. 103:205-217. Hawke, and C. C. Miyada, 1991. Immunoglobu- Soller, M., and J. S. Backmann, 1986. Restriction lin variable-region-like domains of diverse sefragment length polymorphism in poultry quence within the major histocompatibility breeding. Poultry Sci. 65:1474-1488. complex of chicken. Proc. Natl. Acad. Sci. USA Southern, E., 1975. Detection of specific sequences 88:4377-4381. among DNA fragments separated by gel elecMiller, M. M., R. Goto, S. Young, J. Liu, and J. Hardy, trophoresis. J. Mol. Biol. 98:503-517. 1990. Antigens similar to major histocompatibil- Uni, Z., J. Hillel, R. Vaiman, A. Cahaner, and E. D. ity complex B-G are expressed in the intestinal Heller, 1992. Restriction fragment length polyepithelium in the chicken. Immunogenetics 32: morphism of MHC class-rV (B-G) genotypes in 45-50. meat-type chickens. Anim. Genet. 23:319-324.