- Email: [email protected]
A deleted retroviral insertion at the ev21-K complex locus in Indonesian chickens
MOLECULAR BIOLOGY A Deleted Retroviral Insertion at the ev21-K Complex Locus in Indonesian Chickens M. TIXIER-BOICHARD,* A. BOULLIOU-ROBIC,†,1 M. MORISSON,*,1 G. COQUERELLE,* P. HORST,‡ and B. BENKEL§ *Laboratoire de Ge´ne´tique Factorielle, Institut National de la Recherche Agronomique, 78352 Jouy-en-Josas, Cedex, France, †Laboratoire de Ge´ne´tique Mole´culaire, Centre National de la Recherche Scientifique, Unite de Recherche Associe ´e 256, Universite´ de Rennes I, Camus de Beaulieu, 35042 Rennes Cedex, France, ‡Institu ¨ t fur Grundlagen der Nutztierwissenschaften, Humbolt-Universita¨t zu Berlin, Lentzeallee 75 D-14195 Berlin, Germany, and §Centre for Food and Animal Research, Agriculture and Agri-Food Canada, Ottawa, Ontario, Canada, K1A 0C6 type a (URa). A deletion in the 5′ region of the provirus and of the insertion site was indicated by the RFLP analysis and confirmed by a PCR study. Primers were designed in order to amplify a 5′ junction fragment specific to the modified ev21 found in the Nunukan chickens. The sequence of this amplified product showed that the deletion started 652 bp upstream of the insertion site of ev21 and ended within the pol gene of the viral genome. This deletion represents a new allele, OSD, at the ev21 insertion site (locus ALVE21), that appears insufficient to produce a complete virus. Current data do not show a clear causal relationship between OSD and the VLF phenotype. The presence of OSD may be required but is not in itself sufficient to obtain the VLF phenotype. The genetic relationships between OSD and the altered feathering phenotype of Nunukan chickens will be investigated further in families segregating for the VLF phenotype, using the locus-specific PCR test developed as part of this study.
(Key words: polymerase chain reaction, endogenous provirus, late-feathering mutation, deletion, chicken) 1997 Poultry Science 76:733–742
The sex-linked dominant late-feathering mutation (K) is characterized by a reduction in the growth rate of wing and tail feathers, as reviewed by Chambers et al. (1993). This mutation was found to be linked with the endogenous viral gene ev21 of the ALVE family in White Leghorn chickens (Bacon et al., 1988). The tight linkage found between ev21 and the K gene suggested that the retroviral insertion was responsible for the mutated phenotype. The genetic structure of the complex ev21-K locus has been further investigated with the help of a locus-specific probe identifying the ev21 insertion site (Levin and Smith, 1990). Surprisingly, late-feathering birds from various origins carrying ev21 (occupied site: OS) also appeared to carry a duplicated unoccupied insertion site (US) (Smith and Levin, 1991). Furthermore,
INTRODUCTION DNA sequences of retroviral origin can be found in the chicken genome. Several families of endogenous viral (ev) elements have been described. The ALVE family (ev genes) consists of endogenous proviruses that are closely related to avian leukosis viruses. The ALVE insertions can be found in many chicken populations, both commercial and laboratory strains (Crittenden, 1991).
Received for publication July 22, 1996. Accepted for publication December 6, 1996. 1Present address: Laboratoire de Ge ´ ne´tique Cellulaire, Institut National de la Recherche Agronomique, B.P.27 31326 CastanetTolosan, France.
733
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
ABSTRACT Very poor feather development has been observed in chickens of the Nunukan strain, originating from Indonesia. The wing of the newly hatched chick does not show any primary or covert feathers; this phenotype will be referred to as very-late feathering (VLF). As adults, chickens are feathered but tail feathers are short and fragile. An experimental population was set up at the National Institute of Agronomic Research (INRA), Jouy-en-Josas, from one Nunukan male and four Nunukan females. Preliminary observations did not support the hypothesis of a sex-linked dominant mode of inheritance for the VLF phenotype. A restriction fragment length polymorphism (RFLP) study using five restriction enzymes and two probes, RAV-2 and endogenous virus (ev) ev21-int specific for the endogenous viral locus ALVE21, showed the presence of the expected 3′ junction fragments for the ev21 occupied site but failed to reveal the expected 5′ junction fragments for ev21 in Nunukan chickens. The unoccupied site corresponded to the ev21 unoccupied repeat (UR) of
734
TIXIER-BOICHARD ET AL.
MATERIALS AND METHODS
Chickens Fertile eggs from the Nunukan strain were sent from the east Kalimanthan region of Indonesia to the experimental facility of the Technische Universita¨t Berlin (presently Humboldt Universita¨t). An experimental subpopulation was then maintained in Berlin. A sample of fertile eggs was subsequently imported from Berlin to France by the Laboratoire de Ge´ne´tique Factorielle (INRA) and produced a hatch of 14 chicks (Generation 0). All the chicks except one exhibited a peculiar feathering
pattern at hatch, showing neither primary nor covert feathers but only the down (Figure 1a). These chicks were scored VLF. One male and six females were kept until the adult age at INRA, Jouy-en-Josas. In order to produce Generation 1, the Nunukan male was mated to two earlyfeathering (EF) ISA brown females (Figure 2, Family 1), whereas four of the Nunukan females were mated to an unrelated late-feathering (LF) male (Figure 2, Family 2). Generation 2 was produced by mating a VLF male from Family 1 to one VLF female and two LF females from Family 2, as well as to one unrelated early feathering (EF) female (Figure 2). The size of this experimental population was increased in following generations, by crossing VLF females to unrelated EF males, and VLF males to VLF females as well as to unrelated LF or EF females within a brown-egg layer genetic background. Feathering phenotype was scored at hatch in each generation and the VLF phenotype could be regularly observed. Chicks were vent-sexed and reared on the floor together with chicks from other genetic origins. They were provided ad libitum access to a standard grower diet (20% crude protein, 2,900 kcal metabolizable energy) and received 10 h light, 14 h darkness. Feathering condition was recorded again at 8 wk of age. Females were moved to individual cages at 17 wk of age and were inseminated at 1 yr of age to produce the next generation. In Generation 4, a male (No. 575) was identified that produced VLF progeny whatever the feathering phenotype of the dam. This male, apparently homozygous for the VLF condition, was also mated to 5 EF White Leghorn females from the WG line (Gavora et al., 1989), and produced 22 crossbred chicks with the VLF phenotype that were kept for further backcross studies.
Molecular Study of the ev21 Insertion Site This study involved a subset of the experimental population, including 12 VLF male and female chicks and 2 female LF chicks hatched in Generation 1 (Figure 2), and the homozygous male No. 575 with its crossbred progeny. Reference EF and LF samples from commercial strains were obtained from a previous study (Boulliou et al., 1992). RFLP Study. Blood samples were collected in EDTA at 8 wk of age. Genomic DNA from Generation 1 was extracted from 80 mL of blood after hemolysis, incubation with proteinase K, purification with phenol-chloroform and precipitation with isopropanol, as described previously (Tixier-Boichard et al., 1994a). From Generation 3 on, the DNA extraction procedure was simplified and was based upon hemolysis at 4 C in lysis buffer (10 mM Tris pH 8, 1 mM MgCl2, 150 mM NaCl, and 1% NP-40), incubation with extraction buffer (TE-buffer) (50 mM Tris pH 9.5, 100 mM EDTA, 1% SDS, and 200 mg/mL proteinase K) followed by a first precipitation with 4.5 mL dimethylformamide/acetone (5:95 vol/vol) resuspension in TE-buffer and a second precipitation with 100% ethanol. Genomic DNA was finally resuspended in 1 to 2
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
the presence of ev21 without the US repeat could be found in early-feathering birds in various chicken strains (Boulliou et al., 1992). Thus, the presence of ev21 alone is not sufficient to induce the mutant late-feathering phenotype. More recently, the study of the US and OS regions with rate-cutting endonucleases has shown that the OS region was also characterized by a 32-kbp deletion as compared to the US region (Iraqi and Smith, 1995). Thus, the late-feathering mutation seems to be associated with a modification of a genomic region on the Z chromosome, which includes a retroviral insertion. In this paper, we have used another approach to improve our knowledge of the ev21-K complex, based upon the characterization of birds showing a variant phenotype for the rate of feathering. Indeed, alleles other than the late-feathering K allele have been described at the sex-linked locus for rate of feather growth in the chicken (Somes, 1969; McGibbon, 1977). As we could not find a population still carrying any of these alleles, we became interested in the Nunukan strain, a native population from the East Kalimanthan region of Indonesia. Chicks of this strain showed a very slow rate of feather growth, a phenotype that appeared to be of genetic origin. The Nunukan birds were imported to Germany and multiplied as an experimental strain by P. Horst. Fertile eggs of this strain were brought to France, where birds were reproduced in standard environmental conditions. In order to investigate whether this very-late-feathering (VLF) is a mutation at the K locus, experimental crosses were carried out and the molecular analysis of the ev21-K complex was undertaken. In this paper, we report the results of the molecular analysis that revealed the presence of a deletion in the retroviral sequence and in the 5′ flanking region of the ev21 insertion site. We propose to describe this variant by a new allele symbol, OSD, at the ALVE21 locus. In addition, we describe a new PCR diagnostic test for this deleted variant of ev21, that will allow further investigation of the relationship between the deleted ev21 and the modified rate of feathering.
ev21-K LOCUS IN INDONESIAN CHICKENS
735
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
FIGURE 1. Characteristics of the very-late-feathering (VLF) phenotype; top) the wing of a newly hatched VLF chick on the left, the wing of a newly hatched early-feathering (EF) chick on the right; bottom) a VLF male chick at 8 wk of age (Generation 1 of the experimental population maintained at INRA, Jouy-en-Josas).
mL of TE-buffer and its concentration determined by spectrophotometry.
2Applige ` ne, 67402 Illkirch, France; Boehringer Mannheim France S.A., 38242 Meylan, France. 3Pall France, 3 rue de Gaudines BP 253 78104 Saint-Germain-en-Laye Cedex, France.
Aliquots of 10 mg DNA were digested with one of the five restriction enzymes, SacI, EcoRI, HaeIII, HindIII, and PvuII (2 to 5 units per microgram of genomic DNA) under conditions recommended by the supplier.2 Digested DNA samples were fractionated on a 1% agarose gel at 2 V/cm for 18 h. The DNA was then blotted onto a charged nylon membrane (Biodyne B, Pall3) in 0.4 M NaOH. After transfer, the membranes were washed with 2 × SSC and
736
TIXIER-BOICHARD ET AL.
prehybridized for at least 4 h in 5 × SSPE, 1% SDS, 50% formamide, 0.1% Denhardt’s solution, and 200 mg/mL sonicated salmon sperm DNA. The membranes were hybridized with 32P-labeled probes (107 counts per minute for a blot of 400 cm2) for 20 h at 42 C. After hybridization, the membranes were washed twice at room temperature with 2 × SSPE and 0.5% SDS and twice at 65 C with either 0.5 × or 0.1 × SSPE. The membranes were blotted dry and autoradiographed at –80 C using either Kodak Safety film AR or Fuji film and intensifying screens. Three probes were used. The plasmid pRAV-2 is a pBR322 plasmid containing cDNA sequences of genes gag, pol, env, and long terminal repeats (LTR) (Smith and Crittenden, 1986). The RAV-2 sequences (7.6 kbp) were separated from the plasmid by SalI digestion. An env subprobe was obtained by KpnI and SalI digestion of the pRAV-2 clone (Boulliou et al., 1991). The ev21-int probe is a 1.6-kbp genomic fragment flanking the insertion site of ev21 that was cloned in pUC18 by Levin and Smith (1990). The ev21-int sequence was also separated by an EcoRI digestion of the plasmid previously linearized with
4Amersham France SA, 91944 Les Ulis, France. 5Eurogentec, CHRU Ba ˆ t. Monte´clair, 49033 Angers Cedex, France. 6Applied Biosystems Canada Inc., Mississauga, ON, Canada, L5N
2M2.
BamHI. Probes were labeled with a random-primed DNA labelling kit.4 The restriction fragments that are expected without the presence of the ev21 provirus (US locus) or with the ev21 provirus (ALVE21*OS allele) were characterized previously with RAV-2 and ev21-int probes (Smith and Levin, 1991; Boulliou et al., 1992). PCR Study. The PCR analysis was conducted in two stages. Firstly, preliminary PCR tests were carried out to verify the RFLP results. This stage culminated in the amplification of the upstream junction fragment of the modified allele found at the ALVE21 locus in Nunukan chickens. Next, a locus-specific PCR test was developed for the ALVE21 locus that distinguishes between the wellknown OS allele and the new allele found in Nunukan chickens, hereafter named OSD. Primers were designed according to genomic and viral sequences published previously (Levin and Smith, 1991; Ronfort et al., 1991; Schwartz et al., 1993). The approximate priming locations for all primers are diagrammed in Figure 3. Oligonucleotides were either ordered from Eurogentec5 or synthesized on an Applied Biosystems (ABI6) synthesizer Model 381A11 and deblocked and desalted before use. Initially, high molecular weight DNA extracted as described above was employed as substrate for PCR reactions. From Generation 3 onward, crude DNA extracts for PCR assay
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
FIGURE 2. Origin and mating scheme of experimental animals from generations 0 (G0) to 4 (G4). The square symbol represents male chickens and the circle, females. The black squares or circles correspond to VLF chickens, the grey ones to LF chickens and the open ones to EF chickens. The numbers indicate how many chickens of each phenotype were obtained in a given family. The introduction of unrelated animals from a brown-egg layer background is indicated by an arrow.
ev21-K LOCUS IN INDONESIAN CHICKENS
737
FIGURE 3. Schematic representation of the insertion site of endogenous virus (ev) ev21 showing the annealing regions for the primers used in the PCR typing strategy. Primers PR-UP, PR-DWN and P1-1N have been used previously for the basicev21 locus-specific test (Tixier-Boichard et al., 1994b). P2-UP is a new oligomer that primes at position 1505 in the USIL sequence (EMBL accession number X54093) 706 bp upstream of theev21 insertion site. PN-UP primes 8 bases downstream of PR-UP and was chosen to improve the quality of the PCR test for routine typings. Primers PR-GAG, POL-2, and ENVU are designed to anneal 143 bp downstream of the 5′ LTR in the leader region of the provirus, in the pol gene of the proviral element at position 4615, and at the beginning of the env gene at position 5266 (Schwartz et al., 1983), respectively.
A Locus-Specific PCR Test for the ALVE21*OSD Allele
Preliminary PCR Analysis and DNA Sequencing
The PCR test for the OSD allele consists of a modified and expanded version of the assay for ev21. In order to accommodate the number and variety of products and to reduce primer interference in multiplex reactions, the OSD assay was performed in two reaction tubes, side-byside, using different primer combinations. Reaction A combines PR-DWN and P1-1N, the downstream and LTRspecific primers from the ev21 tests (Tixier-Boichard et al., 1994b) with a new primer for the upstream flanking region, PN-UP (5′-GAGATGTCTTCATGCAAATAG-3′). This assay determines the overall status of the bird for the US site and the OS site in its 3′ flanking region. Reaction B combines primer PN-UP with PR-GAG, and POL-2 (5′CTCGCGTGGATTTAGACGTGA-3′), which primes within the pol gene of the virus and is based on the DNA sequence derived in this study for the OSD allele. This reaction discriminates between complete (OS) and 5′-deleted proviral elements (OSD) at the ev21 insertion site. The sequences of the oligomers used in the PCR assays were optimized using OLIGO software.8 The PCR reactions were carried out under standard conditions in 50 mL volumes using a Perkin Elmer Cetus model 9600 thermocycler. Approximately 50 pmol of each primer were used per reaction, except for primer P1-1N for which the amount was raised to 150 pmol. In addition, the final magnesium concentration was raised to 3 mM. A modified “touchdown” procedure was used in order to minimize secondary priming (Don et al., 1991; Benkel and Smith, 1993). The PCR profile consisted of 96 C for 2 min; 2 cycles of 96 C for 45 s, 61 C for 30 s, 72 C for 2 min; two cycles of 96 C for 45 s, 58 C for 30 s, 72 C for 2 min; two cycles of 94 C for 45 s, 55 C for 30 s, 72 C for 2 min; 28 cycles of 94 C for 45 s, 52 C for 30 s, 72 C for 2 min; followed by a 10 min extension at 72 C. Amplified products were separated on a 1.5% agarose gel containing ethidium bromide.
Preliminary PCR amplifications were carried out using standard ev21 test primers PR-UP (5′GTGGGAATGGTACTACAGAGAAGG-3′), P1-1N (5′ACCTGAATGAAGCTGAAGGCTTC-3′) and PR-DWN (5′-CATTTCAAGCAAGGGACTGGC-3′) as described in Tixier-Boichard et al. (1994b). In addition, three new primers were used to further characterize the 5′ flanking region of the OSD allele in Nunukan chickens. These primers were PR-GAG (5′-ACTCAGCTTCTGCCCTCCTAAGCC-3′), which anneals just downstream of the 5′ LTR, P2-UP (5′-GACATCTTGAGATGTCTTCATGC-3′) and ENVU (5′-CTGCTCGAGTAAGTGGACATC-3′). The PCR amplification was performed with primers P2UP and ENVU under the conditions described for the ev21 PCR test (Tixier-Boichard et al., 1994b). The amplified product from a VLF chicken was sequenced in order to precisely determine the structure of the 5′ end of the provirus and the upstream flanking region. The sequencing was carried out on double-stranded DNA using the Applied Biosystems Inc. (ABI) dyedeoxyterminator cycle sequencing kit according to the instruction supplied by the manufacturer. Extension reactions were analyzed on an ABI automated sequencer Model 373A. Amplified PCR products were either sequenced directly, or were subcloned into the pCRII vector (Invitrogen7) prior to sequence analysis. The sequence was compared to the previously published sequences for the ev21 upstream flanking region (Levin and Smith, 1991) and the Rous Sarcoma Virus (Schwartz et al., 1983).
7Invitrogen, De Schelp 12 9351 8National Biosciences, Plymouth
NV Leek, The Netherlands. MN 55447.
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
were prepared according to the Protocol B of Higuchi (1989) as described in Benkel et al. (1992).
738
TIXIER-BOICHARD ET AL. TABLE 1. Size of restriction fragments revealed after hybridization with endogenous virus (ev) ev21-int probe of digestions of genomic DNA from chickens of various feathering phenotypes1
Animals studied
Locus
EcoRI digest
HaeIII digest
HindIII digest
PvuII digest
2.5 2.9 2.5 or3 2 3.5/2.9 2
1.9 0.9 1.9 7.6/0.9 1.9
(kbp) VLF and LF Nunukan chicks LF reference samples EF reference samples
US OS US OS US
3.5 2.5 3.5 4.7/2.5 3.5
2.52 2.45 2.5 or3 1.6 2.45/0.9 2.5 or3 1.6
1EF
= early feathering; LF = late feathering; VLF = very late feathering. chickens exhibited the URa allele (Smith and Levin, 1991). 3According to the genetic origin of the sample (brown-egg layer, White Leghorn, or commercial broiler), polymorphisms were observed at the US locus, in agreement with previous data (Boulliou et al., 1992). 2VLF
RESULTS AND DISCUSSION
Except for one chick that was EF, the founder chicks (Generation 0) received initially in Jouy-en-Josas exhibited the new VLF phenotype at hatch. The VLF chicks were still poorly feathered at 8 wk of age, showing generally very few feathers on the wings and the back (Figure 1b). This phenotype is similar to the description of the delayed feathering allele, Kn, dominant over K, reported by Somes (1969). However, the feathering condition of VLF chicks at 8 wk of age was found to vary between contemporary families produced in subsequent generations. This can be due to additive genetic variation that is known to occur for the degree of feathering in chicken strains, either EF or LF, as reviewed by Chambers et al. (1993). At the adult age, VLF birds were almost normally feathered, but the feathers appeared to be of poor quality, being easily broken and sometimes curly. The tail feathers of the adults were very short and fragile in both sexes. Although the aim of this paper was not to analyze the inheritance of the VLF trait, a few preliminary observations deserve consideration. In Family 1 of Generation 1, 5 females and 5 males among the 12 chicks from the VLF founder male exhibited the VLF phenotype (Figure 2), but one male and one female showed primary feathers that were shorter than covert feathers that corresponded to the LF phenotype. This finding suggested that Generation 0 of this experimental population (Nunukan chickens) was heterogenous with regard to genetic determination of the rate of feathering. In Family 2 of Generation 1, 3 VLF female chicks were found in the progeny of one Nunukan female mated to an unrelated LF sire (Figure 2). This transmission pattern was not in agreement with the hypothesis of a sex-linked mode of inheritance. Only one of these three females survived but unfortunately did not produce any chicks. In later generations, both LF and VLF chicks were regularly obtained from VLF dams, whatever the genotype of the sire, which was not in agreement with the hypothesis of a sex-linked dominant mode of inheritance and also not consistent with the hypothesis of
RFLP Analysis The use of the ev21-int probe revealed the restriction fragments expected with EcoRI, HaeIII, HindIII, or PvuII for the US region and for the 3′ flanking region of the OS allele but failed to show the 5′ flanking fragments of OS in the VLF chickens (Table 1; Figure 4). The OS 5′ flanking fragments were also absent in the two LF chicks hatched in Family 1 of Generation 1. These observations suggested the presence in Nunukan chickens of a retroviral insertion at the ev21 insertion site with a deletion in the 5′ region of the OS. Moreover, the hybridization of the SacI digests with ev21-int probe revealed the presence of a 10-kbp fragment typical for the US locus in all types of chickens, but the other fragments clearly differed according to the individuals studied. The reference LF chickens showed two fragments of 9 and 8.5 kbp corresponding to the OS allele while the Nunukan chickens lacked these bands and exhibited one new fragment of 13 kbp (Figure 4b). Furthermore, the RAV-2 probe hybridized only with the 9-kbp fragment of LF chickens (3′ flanking fragment of OS) and with the 13-kbp fragment of Nunukan chickens (data not shown). These results suggested that the SacI site usually found in the 5′ region of the provirus was missing. The region between the two genomic SacI sites flanking the ev21 provirus is estimated to be 17.5 kbp long (9 + 8.5). The size of the deletion suggested in the Nunukan chickens would thus be roughly 4.5 kbp (17.5 to 13). The use of the env subprobe on HindIII and EcoRI digests provided additional information on the structure of the inserted provirus. Each of these two enzymes recognizes two sites within a complete provirus, thereafter named upstream and downstream (Shank et al., 1981), whatever the location of the provirus in the chicken genome. The env subprobe should hybridize with this internal viral fragment (3.5 kbp after HindIII digestion and 3.9 kbp after
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
Feathering Phenotype
a dominant allele at the sex-linked K locus, such as Kn. At that stage, it was hoped that the molecular study could be used to describe the ev21-K complex locus in the VLF chickens from the Nunukan strain and either clarify or rule out any possible relationships of the VLF phenotype with the ev21-K complex locus.
ev21-K LOCUS IN INDONESIAN CHICKENS
739
EcoRI digestion). Nunukan chickens were known to harbor several uncharacterized ev loci revealed by the hybridization of the RAV-2 probe on SacI digests (data not shown). Consequently, the expected internal fragments were revealed with the env subprobe, that can be also due to the presence of other ev loci. However, among the different types of chickens studied, Nunukan chickens were the only individuals to exhibit a new HindIII fragment of 2.1 kbp and a new EcoRI fragment of 3.5 kbp. These fragments were considered to be the 5′ flanking fragments of a deleted provirus that had lost the upstream internal site for either HindIII or EcoRI but had kept at
least a part of the env gene with the downstream sites for HindIII and EcoRI. For the OS allele (occupied insertion site at the ALVE21 locus), the distance between the HindIII site located in the 5′ flanking genomic region and the most downstream HindIII site of the provirus is 7 kbp, which differs by 4.9 kbp from the size of the HindIII fragment found in Nunukan chickens. The corresponding figure for the EcoRI sites is 8.6 kbp for the OS allele, which differs by 5.1 kbp from the size of the EcoRI fragment found in Nunukan chickens. The different combinations of probes and restriction enzymes were thus consistent with the hypothesis of a deletion covering 4.5 to 5 kbp, starting
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
FIGURE 4. Restriction Fragment Length Polymorphism study of endogenous virus (ev) ev21 provirus and its insertion site. a) restriction maps of the unoccupied (US) and occupied (OS) regions, showing the restriction sites for SacI (S), EcoRI (E), and HindIII (H), the hatched box corresponds to the deletion identified in Nunukan chickens. b) autoradiogram of aSacI blot hybridized with the ev21-int probe, Lane 1: ladder; Lane 2: reference EF sample; Lanes 3 and 5: reference LF samples; Lane 4: VLF sample from Nunukan origin; the 10.4-kbp fragment corresponds to the unoccupied site of ev21, the 9 kbp fragment to the 3′ junction fragment of ev21, the 8.5-kbp fragment to the 5′ junction fragment of ev21; the 13 kbp is a new fragment observed only in the very late feathering (VLF) chicken, which showed also the 10.4 kbp; c) autoradiogram of an EcoRI blot hybridized with the ev21-int probe, Lane 1: ladder; Lane 2: Sire 575; Lanes 3–5 and 7: VLF crossbred progeny; Lane 6: early feathering WG dam. The upper band of 3.5 kbp corresponds to the unoccupied site of ev21, the lower band of 2.5 kbp corresponds to the 3′ junction fragment of ev21.
740
TIXIER-BOICHARD ET AL.
after the HindIII site in the upstream flanking region and including the 5′ region of the provirus. This information was used to prepare the P2-UP primer for the PCR analysis, which was designed to anneal upstream of the conserved HindIII site.
PCR Study
TABLE 2. Haplotypes at the endogenous virus (ev) ev21 insertion site and their corresponding PCR reaction products, found in chickens of various feathering phenotypes1 PCR reaction A (PN-UP/P1-1N/PR-DWN)
PCR reaction B (PN-UP/PR-GAG/POL-2)
Animal studied
Locus
PN-UP/PR-DWN 1,060 bp
P1-1N/PR-DWN 390 bp
PN-UP/PR-GAG 1,115 bp
PN-UP/POL-2 600 bp
EF LF VLF EF
US only US+OS US+OSD OS only
+ + + –
– + + +
– + – +
– – + –
1EF
= early feathering; LF = late feathering; VLF = very late feathering.
2Although the priming sites for both primers are present in the genome of these birds, the amplified product should be greater than
not observed under standard PCR conditions.
5 kbp, which is
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
The PCR study involved only the Nunukan chicks showing the VLF phenotype and reference LF and EF samples. In preliminary PCR analyses, we observed that the standard ev21 PCR test with primers PR-UP, P1-1N and PR-DWN (Tixier-Boichard et al., 1994b) yielded the two fragments expected in LF and VLF chickens—a short fragment corresponding to the 3′ end of the OS allele and a longer fragment from the US region. Also, as expected, EF chickens showed only the longer fragment corresponding to the US locus (data not shown). In addition, the USspecific primer combination of P2-UP (which anneals upstream of PR-UP) with PR-DWN resulted in an amplified band of over 1,000 bp in standard EF, LF, and VLF chickens, as predicted from the sequence information for this locus (Levin and Smith, 1991). Taken together, these data indicated that the US and the 3′ end of the OS regions of VLF chickens were indistinguishable from their counterparts in LF chickens. In contrast, the combination of PR-UP and PR-GAG yielded a fragment of over 1,100 bp in LF birds only; no amplified product could be obtained with these primers in VLF or standard EF chickens. Moreover, the combination of PR-UP with ENVU did not yield any amplified product whatever the type of chicken studied, whereas P2-UP and ENVU, which were expected to flank the upstream and downstream ends of the deletion according to the RFLP data, yielded a product of approximately 1,300 bp on VLF chickens only (data not shown). These results were consistent with the RFLP mapping of the OSD allele in VLF birds, which indicated a sizable deletion involving the 5′ end of the ev21 proviral element and hundreds of base pairs of the upstream flanking region. The P2-UP to ENVU PCR fragment from one VLF bird was sequenced in order to precisely determine the endpoints of the deletion in the OSD locus. Figure 5 shows
that the 5′ end of the deletion lies 652 bp upstream of the ev21 insertion site, at position 1559 of the USIL sequence (Levin and Smith, 1991). The size of the deletion in the 5′ end of the provirus is more difficult to determine with precision as the entire nucleotide sequence of the endogenous ALVE-type provirus has not been published. However, the downstream end of the deletion corresponds to position 4033 of the sequence published for RSV (Schwartz et al., 1983). If one adds to this the length of the U3 region of the endogenous LTR (175 bp), and assumes that the 5′ sequences of the endogenous provirus and RSV are roughly colinear, then one arrives at an estimate for the size of the deletion in the ev21 provirus of about 4,200 bp. Consequently, the OSD provirus has lost the 5′ LTR, the entire gag region and about half of the pol gene. Our estimate for the size of the entire deletion, including upstream (652 bp) and proviral (4,200 bp) sequences is roughly 4,850 bp. The locus-specific PCR test for the OSD allele was developed to facilitate the typing of large numbers of chickens as part of the ongoing analysis of the association of the OSD genotype with the VLF phenotype. The test consists of two parallel reactions, each containing a triplet of oligomer primers (Table 2). Component A of the test is a modification of the standard ev21 assay (Tixier-Boichard et al., 1994b) that substitutes a new upstream primer, PN-UP, for PR-UP to compensate for the deletion of a portion of the upstream flanking region in OSD. This reaction establishes the overall status of the bird with respect to the US and OS regions, but cannot distinguish between LF and VLF birds. In this test, standard EF chickens (US only) show a single band of 1,060 bp, resulting from the interaction of primers PN-UP and PR-DWN (Figure 6, Lane 1). In contrast, EF birds carrying the OS allele only (Boulliou et al., 1992) show a single band of 390 bp based on the P1-1N and PR-DWN interaction (Figure 6, Lane 4). Individuals carrying the US+OS (usually LF) and the US+OSD (derived from the VLF chickens) haplotypes show both the 1,060 bp and the 390 bp amplified fragments (Figure 6, Lanes 2 and 3). The difference between the normal OS and the OSD alleles is specific to the 5′ portion of the proviral insert. Component B of the assay targets this region with primers
ev21-K LOCUS IN INDONESIAN CHICKENS
741
FIGURE 5. Sequence analysis of the OSD allele. The upper line represents the nucleotide sequence of a continuous section of the P2-UP to ENVU PCR fragment from the very late feathering (VFL) chicken. The lower line represents sequence information from two different sources. The 5′ end corresponds to the upstream flanking sequence of the endogenous virus (ev) ev21 insertion site (Levin and Smith, 1991). The conserved HindIII endonuclease site is shown in italics, and the deletion endpoint in the OSD allele at 653 bp upstream of the ev21 insertion site is marked by ˆ 5′. The 3′ end of the lower line shows the sequence of Rous Sarcoma Virus (Schwartz et al., 1983). The arrow, ˆ 3′, marks position 4033, roughly in the middle of the pol of RSV. The motif “GAAA”, which is underlined in the top line of the figure, is present only once in the OSD sequence. It is shown here at either end of the deletion due to the fact that it is not possible to determine whether the GAAA in the OSD allele sequence was derived from the upstream flanking region or the pol gene of the provirus. The deletion in the OSD allele is 4,850 bp long, including 652 bp of upstream flanking region, the entire upstream long terminal repeat (LTR) and gag region of the provirus, plus roughly half of the pol coding region.
insertion site was found in all VLF chickens but also in 2 LF chickens from Generation 1. The VLF phenotype appears to be associated with OSD because of a founder effect. Indeed, only one Nunukan male was at the origin of the experimental subpopulation studied at INRA, Jouyen-Josas. This male was probably homozygous for OSD but not for the VLF condition. The male, No. 575, which was found in Generation 4 to be homozygous for the VLF condition, was also homozygous for OSD. This male’s crossbred progeny, born from EF dams, were VLF and carried OSD (Figure 4c). This result shows that the VLF phenotype was transmitted in a dominant way by Sire
Consequences of the Deletion at the ev21-K Locus The deletion described in the present study represents a new allele at the ev21 insertion site, which we have named OSD. The 3′ flanking region of OSD seems to be identical to the 3′ flanking region of the OS allele previously described in LF chickens. It seems likely that the deletion in the provirus and the 5′ flanking region occurred after the insertion of a complete provirus. It is not known whether the OSD allele of VLF chickens also carries the 32 kbp deletion described for the OS allele of LF chickens by Iraqi and Smith (1995). Obviously, the retroviral insertion will not be able to lead to the production of a complete viral particle, but could still produce the env protein, as observed with ev6 (Robinson et al., 1981). This point will have to be investigated by a search for RNA expression and the use of an in vitro system that could reveal the presence of the env protein. The important question to address is whether the deletion is responsible for the altered feathering phenotype VLF. The limited number of birds and the sampling strategy used in the present study does not allow for linkage analysis. However, our preliminary information on the inheritance of the VLF phenotype does not support the hypothesis of a sex-linked dominant allele to the K mutation. The structural modification observed at the ev21
FIGURE 6. A PCR test that discriminates between the OS and OSD alleles. A typical 1.5% agarose gel showing the amplified PCR products from reactions A (Lanes 1 to 4) and B (Lanes 5 to 8) of the OSD specific assay. Reaction A contains primers that are specific to the unoccupied endogenous virus (ev) ev21 integration site (US) and the 3′ end of the occupied site (OS). Reaction B discriminates between the OS and OSD alleles. Reference early feathering (EF) chickens carrying the US only yield a band of 1,060 bp in reaction A (Lane 1), and no product in reaction B (Lane 5); late feathering (LF) chickens carrying US+OS haplotype yield the 1,060-bp band (specific to the US) plus a 390-bp band derived from the 3′ end of the OS allele in reaction A (Lane 2), and the 1,115-bp upstream junction fragment in reaction B (Lane 6); very late feathering (VLF) chickens carrying the US+OSD haplotype display an identical pattern to the LF (US+OS) for reaction A (Lane 3), but show a single 600-bp upstream junction fragment in reaction B (Lane 7); EF chickens carrying the OS only yield a single band of 390-bp in reaction A (Lane 4), and the 1,115-bp upstream junction band in reaction B (Lane 8). For details of the PCR assay, see text, Figure 3 and Table 2.
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
PR-GAG and POL-2. Standard EF chickens are free of ev21 and do not show any amplified product in this assay (Figure 6, Lane 5). In chickens carrying the OSD allele, primers PN-UP and POL-2 combine to yield an amplified band of 600 bp (Figure 6, Lane 7). The LF birds (OS+US), and EF birds that carry the OS only, show a band of 1,115 bp due to the interaction of PN-UP and PR-GAG (Figure 6, Lanes 6 and 8). The interaction of PN-UP with POL-2 on DNA from birds carrying the complete ev21 insert would result in a band of over 6 kbp. Fragments of this size do not amplify under standard PCR conditions. This test is fully informative in females but can not determine the haplotypes present in males, owing to the presence of two Z chromosomes.
742
TIXIER-BOICHARD ET AL.
575. Male progeny obtained from VLF breeding females, however, were not always VLF, which invalidates the hypothesis that VLF is determined by one single dominant allele on the Z chromosome. At present, it can be assumed that OSD is required but not sufficient to obtain the VLF phenotype. The PCR diagnostic test developed in the present study is currently being used on a larger number of VLF, LF, and EF chicks from later generations, in order to obtain more data on the relationship between OSD and the VLF phenotype. The crossbred progeny of Sire 575 are being backcrossed to EF females of the WG line, in order to follow the segregation of the feathering phenotype and of OSD.
ACKNOWLEDGMENT The authors are grateful to E. J. Smith, who provided the RAV-2 probe and the ev21-int probe.
Bacon, L. D., E. J. Smith, L. B. Crittenden, and G. B. Havenstein, 1988. Association of the slow-feathering (K) and an endogenous viral (ev21) gene of the Z chromosome of chickens. Poultry Sci. 67:191–197. Benkel, B. F., J. Mucha, and J. S. Gavora, 1992. A new diagnostic method for the detection of endogenous Rousassociated virus-type provirus in chickens. Poultry Sci. 71: 1520–1526. Benkel, B. F., and E. J. Smith, 1993. Research note: A rapid method for the detection of the Rous-associated endogenous solitary long terminal repeat, ev15. Poultry Sci. 72:1601–1605. Boulliou, A., J. P. Le Pennec, G. Hubert, R. Donal, and M. Smiley, 1991. Restriction fragment length polymorphism analysis of endogenous avian leukosis viral loci: determination of frequencies in commercial broiler lines. Poultry Sci. 70:1287–1296. Boulliou, A., J. P. Le Pennec, G. Hubert, R. Donal, and M. Smiley, 1992. The endogenous retroviral ev21 locus in commercial chicken lines and its relationship with the slow-feathering phenotype (K). Poultry Sci. 71:38–46. Chambers, J. R., E. J. Smith, E. A. Dunnington, and P. B. Siegel, 1993. Sex-linked feathering (K,k+) in chickens: a review. Poult. Sci. Rev. 5:97–116. Crittenden, L. B., 1991. Retroviral elements in the genome of the chicken: implications for poultry genetics and breeding. Crit. Rev. Poult. Biol. 3:73–109. Don, R. H., P. T. Cox, B. J. Wainwright, and J. S. Mattick, 1991. “Touchdown” PCR to circumvent spurious priming during gene amplification. Nucl. Acids Res. 19:4008.
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
REFERENCES
Gavora, J. S., U. Ku¨hnlein, and J. L. Spencer, 1989. Absence of endogenous viral genes in an inbred line of Leghorn chickens selected for high egg production and Marek’s disease resistance. J. Anim. Breed. Genet. 106:217–224. Higuchi, R., 1989. Simple and rapid preparation of samples for PCR. Pages 31–38 in: PCR Technology: Principles and Applications for DNA Amplifications. H. A. Erlich, ed. Stockton Press, New York, NY. Iraqi, F., and E. J. Smith, 1995. Organization of the sex-linked late-feathering haplotype in chickens. Anim. Genet. 26: 141–146. Levin, I., and E. J. Smith, 1990. Molecular analysis of endogenous virus ev21 slow-feathering complex of chickens. 1. Cloning of proviral-cell junction fragment and unoccupied integrate site. Poultry Sci. 69:2017–2026. Levin, I., and E. J. Smith, 1991. Association of a chicken repetitive element with the endogenous virus-21 slowfeathering locus. Poultry Sci. 70:1948–1956. McGibbon, W. H., 1977. A sex-linked mutation affecting rate of feathering in chickens. Poultry Sci. 56:872–875. Robinson, H. L., S. M. Astrin, A. M. Senior, and F. H. Salazar, 1981. Host susceptibility to endogenous viruses: defective glycoprotein-expressing proviruses interfere with infections. J. Virol. 42:745–751. Ronfort, C., M. Afanassieff, Y. Chebloune, G. Dambrine, V. M. Nigon, and G. Verdier, 1991. Identification and structure analysis of endogenous proviral sequences in a Brown Leghorn chicken strain. Poultry Sci. 70:2161–2175. Schwartz, D. E., R. Tizard, and W. Gilbert, 1983. Nucleotide sequence of Rous sarcoma virus. Cell 32:853–869. Shank, P. R., S. H. Hughes, and H. E. Varmus, 1981. Restriction endonuclease mapping of the DNA of Rous-associated virus-0 reveals extensive homology in structure and sequence with Avian Sarcoma Virus DNA. Virology 108: 177–188. Smith, E. J., and L. B. Crittenden, 1986. Endogenous virus genes in a slow-feathering line of White Leghorns. Avian Pathol. 15:395–406. Smith, E. J., and I. Levin, 1991. Application of a locus-specific DNA hybridization probe in the analysis of the slowfeathering endogenous virus complex of chickens. Poultry Sci. 70:1957–1964. Somes, R. G., Jr., 1969. Delayed feathering, a third allele at the K locus of the domestic fowl. J. Hered. 60:281–286. Tixier-Boichard, M., L. Durand, M. Morisson, F. Ricard, and G. Coquerelle, 1994a. Comparative analysis of avian leukosisrelated endogenous viral genes in experimental strains of the domestic chicken. Genet. Sel. Evol. 26:53s–66s. Tixier-Boichard, M., B. F. Benkel, J. R. Chambers, and J. S. Gavora, 1994b. Screening chickens for endogenous virus ev21 viral element by the polymerase chain reaction. Poultry Sci. 73:1612–1616.
(Key words: polymerase chain reaction, endogenous provirus, late-feathering mutation, deletion, chicken) 1997 Poultry Science 76:733–742
The sex-linked dominant late-feathering mutation (K) is characterized by a reduction in the growth rate of wing and tail feathers, as reviewed by Chambers et al. (1993). This mutation was found to be linked with the endogenous viral gene ev21 of the ALVE family in White Leghorn chickens (Bacon et al., 1988). The tight linkage found between ev21 and the K gene suggested that the retroviral insertion was responsible for the mutated phenotype. The genetic structure of the complex ev21-K locus has been further investigated with the help of a locus-specific probe identifying the ev21 insertion site (Levin and Smith, 1990). Surprisingly, late-feathering birds from various origins carrying ev21 (occupied site: OS) also appeared to carry a duplicated unoccupied insertion site (US) (Smith and Levin, 1991). Furthermore,
INTRODUCTION DNA sequences of retroviral origin can be found in the chicken genome. Several families of endogenous viral (ev) elements have been described. The ALVE family (ev genes) consists of endogenous proviruses that are closely related to avian leukosis viruses. The ALVE insertions can be found in many chicken populations, both commercial and laboratory strains (Crittenden, 1991).
Received for publication July 22, 1996. Accepted for publication December 6, 1996. 1Present address: Laboratoire de Ge ´ ne´tique Cellulaire, Institut National de la Recherche Agronomique, B.P.27 31326 CastanetTolosan, France.
733
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
ABSTRACT Very poor feather development has been observed in chickens of the Nunukan strain, originating from Indonesia. The wing of the newly hatched chick does not show any primary or covert feathers; this phenotype will be referred to as very-late feathering (VLF). As adults, chickens are feathered but tail feathers are short and fragile. An experimental population was set up at the National Institute of Agronomic Research (INRA), Jouy-en-Josas, from one Nunukan male and four Nunukan females. Preliminary observations did not support the hypothesis of a sex-linked dominant mode of inheritance for the VLF phenotype. A restriction fragment length polymorphism (RFLP) study using five restriction enzymes and two probes, RAV-2 and endogenous virus (ev) ev21-int specific for the endogenous viral locus ALVE21, showed the presence of the expected 3′ junction fragments for the ev21 occupied site but failed to reveal the expected 5′ junction fragments for ev21 in Nunukan chickens. The unoccupied site corresponded to the ev21 unoccupied repeat (UR) of
734
TIXIER-BOICHARD ET AL.
MATERIALS AND METHODS
Chickens Fertile eggs from the Nunukan strain were sent from the east Kalimanthan region of Indonesia to the experimental facility of the Technische Universita¨t Berlin (presently Humboldt Universita¨t). An experimental subpopulation was then maintained in Berlin. A sample of fertile eggs was subsequently imported from Berlin to France by the Laboratoire de Ge´ne´tique Factorielle (INRA) and produced a hatch of 14 chicks (Generation 0). All the chicks except one exhibited a peculiar feathering
pattern at hatch, showing neither primary nor covert feathers but only the down (Figure 1a). These chicks were scored VLF. One male and six females were kept until the adult age at INRA, Jouy-en-Josas. In order to produce Generation 1, the Nunukan male was mated to two earlyfeathering (EF) ISA brown females (Figure 2, Family 1), whereas four of the Nunukan females were mated to an unrelated late-feathering (LF) male (Figure 2, Family 2). Generation 2 was produced by mating a VLF male from Family 1 to one VLF female and two LF females from Family 2, as well as to one unrelated early feathering (EF) female (Figure 2). The size of this experimental population was increased in following generations, by crossing VLF females to unrelated EF males, and VLF males to VLF females as well as to unrelated LF or EF females within a brown-egg layer genetic background. Feathering phenotype was scored at hatch in each generation and the VLF phenotype could be regularly observed. Chicks were vent-sexed and reared on the floor together with chicks from other genetic origins. They were provided ad libitum access to a standard grower diet (20% crude protein, 2,900 kcal metabolizable energy) and received 10 h light, 14 h darkness. Feathering condition was recorded again at 8 wk of age. Females were moved to individual cages at 17 wk of age and were inseminated at 1 yr of age to produce the next generation. In Generation 4, a male (No. 575) was identified that produced VLF progeny whatever the feathering phenotype of the dam. This male, apparently homozygous for the VLF condition, was also mated to 5 EF White Leghorn females from the WG line (Gavora et al., 1989), and produced 22 crossbred chicks with the VLF phenotype that were kept for further backcross studies.
Molecular Study of the ev21 Insertion Site This study involved a subset of the experimental population, including 12 VLF male and female chicks and 2 female LF chicks hatched in Generation 1 (Figure 2), and the homozygous male No. 575 with its crossbred progeny. Reference EF and LF samples from commercial strains were obtained from a previous study (Boulliou et al., 1992). RFLP Study. Blood samples were collected in EDTA at 8 wk of age. Genomic DNA from Generation 1 was extracted from 80 mL of blood after hemolysis, incubation with proteinase K, purification with phenol-chloroform and precipitation with isopropanol, as described previously (Tixier-Boichard et al., 1994a). From Generation 3 on, the DNA extraction procedure was simplified and was based upon hemolysis at 4 C in lysis buffer (10 mM Tris pH 8, 1 mM MgCl2, 150 mM NaCl, and 1% NP-40), incubation with extraction buffer (TE-buffer) (50 mM Tris pH 9.5, 100 mM EDTA, 1% SDS, and 200 mg/mL proteinase K) followed by a first precipitation with 4.5 mL dimethylformamide/acetone (5:95 vol/vol) resuspension in TE-buffer and a second precipitation with 100% ethanol. Genomic DNA was finally resuspended in 1 to 2
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
the presence of ev21 without the US repeat could be found in early-feathering birds in various chicken strains (Boulliou et al., 1992). Thus, the presence of ev21 alone is not sufficient to induce the mutant late-feathering phenotype. More recently, the study of the US and OS regions with rate-cutting endonucleases has shown that the OS region was also characterized by a 32-kbp deletion as compared to the US region (Iraqi and Smith, 1995). Thus, the late-feathering mutation seems to be associated with a modification of a genomic region on the Z chromosome, which includes a retroviral insertion. In this paper, we have used another approach to improve our knowledge of the ev21-K complex, based upon the characterization of birds showing a variant phenotype for the rate of feathering. Indeed, alleles other than the late-feathering K allele have been described at the sex-linked locus for rate of feather growth in the chicken (Somes, 1969; McGibbon, 1977). As we could not find a population still carrying any of these alleles, we became interested in the Nunukan strain, a native population from the East Kalimanthan region of Indonesia. Chicks of this strain showed a very slow rate of feather growth, a phenotype that appeared to be of genetic origin. The Nunukan birds were imported to Germany and multiplied as an experimental strain by P. Horst. Fertile eggs of this strain were brought to France, where birds were reproduced in standard environmental conditions. In order to investigate whether this very-late-feathering (VLF) is a mutation at the K locus, experimental crosses were carried out and the molecular analysis of the ev21-K complex was undertaken. In this paper, we report the results of the molecular analysis that revealed the presence of a deletion in the retroviral sequence and in the 5′ flanking region of the ev21 insertion site. We propose to describe this variant by a new allele symbol, OSD, at the ALVE21 locus. In addition, we describe a new PCR diagnostic test for this deleted variant of ev21, that will allow further investigation of the relationship between the deleted ev21 and the modified rate of feathering.
ev21-K LOCUS IN INDONESIAN CHICKENS
735
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
FIGURE 1. Characteristics of the very-late-feathering (VLF) phenotype; top) the wing of a newly hatched VLF chick on the left, the wing of a newly hatched early-feathering (EF) chick on the right; bottom) a VLF male chick at 8 wk of age (Generation 1 of the experimental population maintained at INRA, Jouy-en-Josas).
mL of TE-buffer and its concentration determined by spectrophotometry.
2Applige ` ne, 67402 Illkirch, France; Boehringer Mannheim France S.A., 38242 Meylan, France. 3Pall France, 3 rue de Gaudines BP 253 78104 Saint-Germain-en-Laye Cedex, France.
Aliquots of 10 mg DNA were digested with one of the five restriction enzymes, SacI, EcoRI, HaeIII, HindIII, and PvuII (2 to 5 units per microgram of genomic DNA) under conditions recommended by the supplier.2 Digested DNA samples were fractionated on a 1% agarose gel at 2 V/cm for 18 h. The DNA was then blotted onto a charged nylon membrane (Biodyne B, Pall3) in 0.4 M NaOH. After transfer, the membranes were washed with 2 × SSC and
736
TIXIER-BOICHARD ET AL.
prehybridized for at least 4 h in 5 × SSPE, 1% SDS, 50% formamide, 0.1% Denhardt’s solution, and 200 mg/mL sonicated salmon sperm DNA. The membranes were hybridized with 32P-labeled probes (107 counts per minute for a blot of 400 cm2) for 20 h at 42 C. After hybridization, the membranes were washed twice at room temperature with 2 × SSPE and 0.5% SDS and twice at 65 C with either 0.5 × or 0.1 × SSPE. The membranes were blotted dry and autoradiographed at –80 C using either Kodak Safety film AR or Fuji film and intensifying screens. Three probes were used. The plasmid pRAV-2 is a pBR322 plasmid containing cDNA sequences of genes gag, pol, env, and long terminal repeats (LTR) (Smith and Crittenden, 1986). The RAV-2 sequences (7.6 kbp) were separated from the plasmid by SalI digestion. An env subprobe was obtained by KpnI and SalI digestion of the pRAV-2 clone (Boulliou et al., 1991). The ev21-int probe is a 1.6-kbp genomic fragment flanking the insertion site of ev21 that was cloned in pUC18 by Levin and Smith (1990). The ev21-int sequence was also separated by an EcoRI digestion of the plasmid previously linearized with
4Amersham France SA, 91944 Les Ulis, France. 5Eurogentec, CHRU Ba ˆ t. Monte´clair, 49033 Angers Cedex, France. 6Applied Biosystems Canada Inc., Mississauga, ON, Canada, L5N
2M2.
BamHI. Probes were labeled with a random-primed DNA labelling kit.4 The restriction fragments that are expected without the presence of the ev21 provirus (US locus) or with the ev21 provirus (ALVE21*OS allele) were characterized previously with RAV-2 and ev21-int probes (Smith and Levin, 1991; Boulliou et al., 1992). PCR Study. The PCR analysis was conducted in two stages. Firstly, preliminary PCR tests were carried out to verify the RFLP results. This stage culminated in the amplification of the upstream junction fragment of the modified allele found at the ALVE21 locus in Nunukan chickens. Next, a locus-specific PCR test was developed for the ALVE21 locus that distinguishes between the wellknown OS allele and the new allele found in Nunukan chickens, hereafter named OSD. Primers were designed according to genomic and viral sequences published previously (Levin and Smith, 1991; Ronfort et al., 1991; Schwartz et al., 1993). The approximate priming locations for all primers are diagrammed in Figure 3. Oligonucleotides were either ordered from Eurogentec5 or synthesized on an Applied Biosystems (ABI6) synthesizer Model 381A11 and deblocked and desalted before use. Initially, high molecular weight DNA extracted as described above was employed as substrate for PCR reactions. From Generation 3 onward, crude DNA extracts for PCR assay
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
FIGURE 2. Origin and mating scheme of experimental animals from generations 0 (G0) to 4 (G4). The square symbol represents male chickens and the circle, females. The black squares or circles correspond to VLF chickens, the grey ones to LF chickens and the open ones to EF chickens. The numbers indicate how many chickens of each phenotype were obtained in a given family. The introduction of unrelated animals from a brown-egg layer background is indicated by an arrow.
ev21-K LOCUS IN INDONESIAN CHICKENS
737
FIGURE 3. Schematic representation of the insertion site of endogenous virus (ev) ev21 showing the annealing regions for the primers used in the PCR typing strategy. Primers PR-UP, PR-DWN and P1-1N have been used previously for the basicev21 locus-specific test (Tixier-Boichard et al., 1994b). P2-UP is a new oligomer that primes at position 1505 in the USIL sequence (EMBL accession number X54093) 706 bp upstream of theev21 insertion site. PN-UP primes 8 bases downstream of PR-UP and was chosen to improve the quality of the PCR test for routine typings. Primers PR-GAG, POL-2, and ENVU are designed to anneal 143 bp downstream of the 5′ LTR in the leader region of the provirus, in the pol gene of the proviral element at position 4615, and at the beginning of the env gene at position 5266 (Schwartz et al., 1983), respectively.
A Locus-Specific PCR Test for the ALVE21*OSD Allele
Preliminary PCR Analysis and DNA Sequencing
The PCR test for the OSD allele consists of a modified and expanded version of the assay for ev21. In order to accommodate the number and variety of products and to reduce primer interference in multiplex reactions, the OSD assay was performed in two reaction tubes, side-byside, using different primer combinations. Reaction A combines PR-DWN and P1-1N, the downstream and LTRspecific primers from the ev21 tests (Tixier-Boichard et al., 1994b) with a new primer for the upstream flanking region, PN-UP (5′-GAGATGTCTTCATGCAAATAG-3′). This assay determines the overall status of the bird for the US site and the OS site in its 3′ flanking region. Reaction B combines primer PN-UP with PR-GAG, and POL-2 (5′CTCGCGTGGATTTAGACGTGA-3′), which primes within the pol gene of the virus and is based on the DNA sequence derived in this study for the OSD allele. This reaction discriminates between complete (OS) and 5′-deleted proviral elements (OSD) at the ev21 insertion site. The sequences of the oligomers used in the PCR assays were optimized using OLIGO software.8 The PCR reactions were carried out under standard conditions in 50 mL volumes using a Perkin Elmer Cetus model 9600 thermocycler. Approximately 50 pmol of each primer were used per reaction, except for primer P1-1N for which the amount was raised to 150 pmol. In addition, the final magnesium concentration was raised to 3 mM. A modified “touchdown” procedure was used in order to minimize secondary priming (Don et al., 1991; Benkel and Smith, 1993). The PCR profile consisted of 96 C for 2 min; 2 cycles of 96 C for 45 s, 61 C for 30 s, 72 C for 2 min; two cycles of 96 C for 45 s, 58 C for 30 s, 72 C for 2 min; two cycles of 94 C for 45 s, 55 C for 30 s, 72 C for 2 min; 28 cycles of 94 C for 45 s, 52 C for 30 s, 72 C for 2 min; followed by a 10 min extension at 72 C. Amplified products were separated on a 1.5% agarose gel containing ethidium bromide.
Preliminary PCR amplifications were carried out using standard ev21 test primers PR-UP (5′GTGGGAATGGTACTACAGAGAAGG-3′), P1-1N (5′ACCTGAATGAAGCTGAAGGCTTC-3′) and PR-DWN (5′-CATTTCAAGCAAGGGACTGGC-3′) as described in Tixier-Boichard et al. (1994b). In addition, three new primers were used to further characterize the 5′ flanking region of the OSD allele in Nunukan chickens. These primers were PR-GAG (5′-ACTCAGCTTCTGCCCTCCTAAGCC-3′), which anneals just downstream of the 5′ LTR, P2-UP (5′-GACATCTTGAGATGTCTTCATGC-3′) and ENVU (5′-CTGCTCGAGTAAGTGGACATC-3′). The PCR amplification was performed with primers P2UP and ENVU under the conditions described for the ev21 PCR test (Tixier-Boichard et al., 1994b). The amplified product from a VLF chicken was sequenced in order to precisely determine the structure of the 5′ end of the provirus and the upstream flanking region. The sequencing was carried out on double-stranded DNA using the Applied Biosystems Inc. (ABI) dyedeoxyterminator cycle sequencing kit according to the instruction supplied by the manufacturer. Extension reactions were analyzed on an ABI automated sequencer Model 373A. Amplified PCR products were either sequenced directly, or were subcloned into the pCRII vector (Invitrogen7) prior to sequence analysis. The sequence was compared to the previously published sequences for the ev21 upstream flanking region (Levin and Smith, 1991) and the Rous Sarcoma Virus (Schwartz et al., 1983).
7Invitrogen, De Schelp 12 9351 8National Biosciences, Plymouth
NV Leek, The Netherlands. MN 55447.
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
were prepared according to the Protocol B of Higuchi (1989) as described in Benkel et al. (1992).
738
TIXIER-BOICHARD ET AL. TABLE 1. Size of restriction fragments revealed after hybridization with endogenous virus (ev) ev21-int probe of digestions of genomic DNA from chickens of various feathering phenotypes1
Animals studied
Locus
EcoRI digest
HaeIII digest
HindIII digest
PvuII digest
2.5 2.9 2.5 or3 2 3.5/2.9 2
1.9 0.9 1.9 7.6/0.9 1.9
(kbp) VLF and LF Nunukan chicks LF reference samples EF reference samples
US OS US OS US
3.5 2.5 3.5 4.7/2.5 3.5
2.52 2.45 2.5 or3 1.6 2.45/0.9 2.5 or3 1.6
1EF
= early feathering; LF = late feathering; VLF = very late feathering. chickens exhibited the URa allele (Smith and Levin, 1991). 3According to the genetic origin of the sample (brown-egg layer, White Leghorn, or commercial broiler), polymorphisms were observed at the US locus, in agreement with previous data (Boulliou et al., 1992). 2VLF
RESULTS AND DISCUSSION
Except for one chick that was EF, the founder chicks (Generation 0) received initially in Jouy-en-Josas exhibited the new VLF phenotype at hatch. The VLF chicks were still poorly feathered at 8 wk of age, showing generally very few feathers on the wings and the back (Figure 1b). This phenotype is similar to the description of the delayed feathering allele, Kn, dominant over K, reported by Somes (1969). However, the feathering condition of VLF chicks at 8 wk of age was found to vary between contemporary families produced in subsequent generations. This can be due to additive genetic variation that is known to occur for the degree of feathering in chicken strains, either EF or LF, as reviewed by Chambers et al. (1993). At the adult age, VLF birds were almost normally feathered, but the feathers appeared to be of poor quality, being easily broken and sometimes curly. The tail feathers of the adults were very short and fragile in both sexes. Although the aim of this paper was not to analyze the inheritance of the VLF trait, a few preliminary observations deserve consideration. In Family 1 of Generation 1, 5 females and 5 males among the 12 chicks from the VLF founder male exhibited the VLF phenotype (Figure 2), but one male and one female showed primary feathers that were shorter than covert feathers that corresponded to the LF phenotype. This finding suggested that Generation 0 of this experimental population (Nunukan chickens) was heterogenous with regard to genetic determination of the rate of feathering. In Family 2 of Generation 1, 3 VLF female chicks were found in the progeny of one Nunukan female mated to an unrelated LF sire (Figure 2). This transmission pattern was not in agreement with the hypothesis of a sex-linked mode of inheritance. Only one of these three females survived but unfortunately did not produce any chicks. In later generations, both LF and VLF chicks were regularly obtained from VLF dams, whatever the genotype of the sire, which was not in agreement with the hypothesis of a sex-linked dominant mode of inheritance and also not consistent with the hypothesis of
RFLP Analysis The use of the ev21-int probe revealed the restriction fragments expected with EcoRI, HaeIII, HindIII, or PvuII for the US region and for the 3′ flanking region of the OS allele but failed to show the 5′ flanking fragments of OS in the VLF chickens (Table 1; Figure 4). The OS 5′ flanking fragments were also absent in the two LF chicks hatched in Family 1 of Generation 1. These observations suggested the presence in Nunukan chickens of a retroviral insertion at the ev21 insertion site with a deletion in the 5′ region of the OS. Moreover, the hybridization of the SacI digests with ev21-int probe revealed the presence of a 10-kbp fragment typical for the US locus in all types of chickens, but the other fragments clearly differed according to the individuals studied. The reference LF chickens showed two fragments of 9 and 8.5 kbp corresponding to the OS allele while the Nunukan chickens lacked these bands and exhibited one new fragment of 13 kbp (Figure 4b). Furthermore, the RAV-2 probe hybridized only with the 9-kbp fragment of LF chickens (3′ flanking fragment of OS) and with the 13-kbp fragment of Nunukan chickens (data not shown). These results suggested that the SacI site usually found in the 5′ region of the provirus was missing. The region between the two genomic SacI sites flanking the ev21 provirus is estimated to be 17.5 kbp long (9 + 8.5). The size of the deletion suggested in the Nunukan chickens would thus be roughly 4.5 kbp (17.5 to 13). The use of the env subprobe on HindIII and EcoRI digests provided additional information on the structure of the inserted provirus. Each of these two enzymes recognizes two sites within a complete provirus, thereafter named upstream and downstream (Shank et al., 1981), whatever the location of the provirus in the chicken genome. The env subprobe should hybridize with this internal viral fragment (3.5 kbp after HindIII digestion and 3.9 kbp after
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
Feathering Phenotype
a dominant allele at the sex-linked K locus, such as Kn. At that stage, it was hoped that the molecular study could be used to describe the ev21-K complex locus in the VLF chickens from the Nunukan strain and either clarify or rule out any possible relationships of the VLF phenotype with the ev21-K complex locus.
ev21-K LOCUS IN INDONESIAN CHICKENS
739
EcoRI digestion). Nunukan chickens were known to harbor several uncharacterized ev loci revealed by the hybridization of the RAV-2 probe on SacI digests (data not shown). Consequently, the expected internal fragments were revealed with the env subprobe, that can be also due to the presence of other ev loci. However, among the different types of chickens studied, Nunukan chickens were the only individuals to exhibit a new HindIII fragment of 2.1 kbp and a new EcoRI fragment of 3.5 kbp. These fragments were considered to be the 5′ flanking fragments of a deleted provirus that had lost the upstream internal site for either HindIII or EcoRI but had kept at
least a part of the env gene with the downstream sites for HindIII and EcoRI. For the OS allele (occupied insertion site at the ALVE21 locus), the distance between the HindIII site located in the 5′ flanking genomic region and the most downstream HindIII site of the provirus is 7 kbp, which differs by 4.9 kbp from the size of the HindIII fragment found in Nunukan chickens. The corresponding figure for the EcoRI sites is 8.6 kbp for the OS allele, which differs by 5.1 kbp from the size of the EcoRI fragment found in Nunukan chickens. The different combinations of probes and restriction enzymes were thus consistent with the hypothesis of a deletion covering 4.5 to 5 kbp, starting
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
FIGURE 4. Restriction Fragment Length Polymorphism study of endogenous virus (ev) ev21 provirus and its insertion site. a) restriction maps of the unoccupied (US) and occupied (OS) regions, showing the restriction sites for SacI (S), EcoRI (E), and HindIII (H), the hatched box corresponds to the deletion identified in Nunukan chickens. b) autoradiogram of aSacI blot hybridized with the ev21-int probe, Lane 1: ladder; Lane 2: reference EF sample; Lanes 3 and 5: reference LF samples; Lane 4: VLF sample from Nunukan origin; the 10.4-kbp fragment corresponds to the unoccupied site of ev21, the 9 kbp fragment to the 3′ junction fragment of ev21, the 8.5-kbp fragment to the 5′ junction fragment of ev21; the 13 kbp is a new fragment observed only in the very late feathering (VLF) chicken, which showed also the 10.4 kbp; c) autoradiogram of an EcoRI blot hybridized with the ev21-int probe, Lane 1: ladder; Lane 2: Sire 575; Lanes 3–5 and 7: VLF crossbred progeny; Lane 6: early feathering WG dam. The upper band of 3.5 kbp corresponds to the unoccupied site of ev21, the lower band of 2.5 kbp corresponds to the 3′ junction fragment of ev21.
740
TIXIER-BOICHARD ET AL.
after the HindIII site in the upstream flanking region and including the 5′ region of the provirus. This information was used to prepare the P2-UP primer for the PCR analysis, which was designed to anneal upstream of the conserved HindIII site.
PCR Study
TABLE 2. Haplotypes at the endogenous virus (ev) ev21 insertion site and their corresponding PCR reaction products, found in chickens of various feathering phenotypes1 PCR reaction A (PN-UP/P1-1N/PR-DWN)
PCR reaction B (PN-UP/PR-GAG/POL-2)
Animal studied
Locus
PN-UP/PR-DWN 1,060 bp
P1-1N/PR-DWN 390 bp
PN-UP/PR-GAG 1,115 bp
PN-UP/POL-2 600 bp
EF LF VLF EF
US only US+OS US+OSD OS only
+ + + –
– + + +
– + – +
– – + –
1EF
= early feathering; LF = late feathering; VLF = very late feathering.
2Although the priming sites for both primers are present in the genome of these birds, the amplified product should be greater than
not observed under standard PCR conditions.
5 kbp, which is
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
The PCR study involved only the Nunukan chicks showing the VLF phenotype and reference LF and EF samples. In preliminary PCR analyses, we observed that the standard ev21 PCR test with primers PR-UP, P1-1N and PR-DWN (Tixier-Boichard et al., 1994b) yielded the two fragments expected in LF and VLF chickens—a short fragment corresponding to the 3′ end of the OS allele and a longer fragment from the US region. Also, as expected, EF chickens showed only the longer fragment corresponding to the US locus (data not shown). In addition, the USspecific primer combination of P2-UP (which anneals upstream of PR-UP) with PR-DWN resulted in an amplified band of over 1,000 bp in standard EF, LF, and VLF chickens, as predicted from the sequence information for this locus (Levin and Smith, 1991). Taken together, these data indicated that the US and the 3′ end of the OS regions of VLF chickens were indistinguishable from their counterparts in LF chickens. In contrast, the combination of PR-UP and PR-GAG yielded a fragment of over 1,100 bp in LF birds only; no amplified product could be obtained with these primers in VLF or standard EF chickens. Moreover, the combination of PR-UP with ENVU did not yield any amplified product whatever the type of chicken studied, whereas P2-UP and ENVU, which were expected to flank the upstream and downstream ends of the deletion according to the RFLP data, yielded a product of approximately 1,300 bp on VLF chickens only (data not shown). These results were consistent with the RFLP mapping of the OSD allele in VLF birds, which indicated a sizable deletion involving the 5′ end of the ev21 proviral element and hundreds of base pairs of the upstream flanking region. The P2-UP to ENVU PCR fragment from one VLF bird was sequenced in order to precisely determine the endpoints of the deletion in the OSD locus. Figure 5 shows
that the 5′ end of the deletion lies 652 bp upstream of the ev21 insertion site, at position 1559 of the USIL sequence (Levin and Smith, 1991). The size of the deletion in the 5′ end of the provirus is more difficult to determine with precision as the entire nucleotide sequence of the endogenous ALVE-type provirus has not been published. However, the downstream end of the deletion corresponds to position 4033 of the sequence published for RSV (Schwartz et al., 1983). If one adds to this the length of the U3 region of the endogenous LTR (175 bp), and assumes that the 5′ sequences of the endogenous provirus and RSV are roughly colinear, then one arrives at an estimate for the size of the deletion in the ev21 provirus of about 4,200 bp. Consequently, the OSD provirus has lost the 5′ LTR, the entire gag region and about half of the pol gene. Our estimate for the size of the entire deletion, including upstream (652 bp) and proviral (4,200 bp) sequences is roughly 4,850 bp. The locus-specific PCR test for the OSD allele was developed to facilitate the typing of large numbers of chickens as part of the ongoing analysis of the association of the OSD genotype with the VLF phenotype. The test consists of two parallel reactions, each containing a triplet of oligomer primers (Table 2). Component A of the test is a modification of the standard ev21 assay (Tixier-Boichard et al., 1994b) that substitutes a new upstream primer, PN-UP, for PR-UP to compensate for the deletion of a portion of the upstream flanking region in OSD. This reaction establishes the overall status of the bird with respect to the US and OS regions, but cannot distinguish between LF and VLF birds. In this test, standard EF chickens (US only) show a single band of 1,060 bp, resulting from the interaction of primers PN-UP and PR-DWN (Figure 6, Lane 1). In contrast, EF birds carrying the OS allele only (Boulliou et al., 1992) show a single band of 390 bp based on the P1-1N and PR-DWN interaction (Figure 6, Lane 4). Individuals carrying the US+OS (usually LF) and the US+OSD (derived from the VLF chickens) haplotypes show both the 1,060 bp and the 390 bp amplified fragments (Figure 6, Lanes 2 and 3). The difference between the normal OS and the OSD alleles is specific to the 5′ portion of the proviral insert. Component B of the assay targets this region with primers
ev21-K LOCUS IN INDONESIAN CHICKENS
741
FIGURE 5. Sequence analysis of the OSD allele. The upper line represents the nucleotide sequence of a continuous section of the P2-UP to ENVU PCR fragment from the very late feathering (VFL) chicken. The lower line represents sequence information from two different sources. The 5′ end corresponds to the upstream flanking sequence of the endogenous virus (ev) ev21 insertion site (Levin and Smith, 1991). The conserved HindIII endonuclease site is shown in italics, and the deletion endpoint in the OSD allele at 653 bp upstream of the ev21 insertion site is marked by ˆ 5′. The 3′ end of the lower line shows the sequence of Rous Sarcoma Virus (Schwartz et al., 1983). The arrow, ˆ 3′, marks position 4033, roughly in the middle of the pol of RSV. The motif “GAAA”, which is underlined in the top line of the figure, is present only once in the OSD sequence. It is shown here at either end of the deletion due to the fact that it is not possible to determine whether the GAAA in the OSD allele sequence was derived from the upstream flanking region or the pol gene of the provirus. The deletion in the OSD allele is 4,850 bp long, including 652 bp of upstream flanking region, the entire upstream long terminal repeat (LTR) and gag region of the provirus, plus roughly half of the pol coding region.
insertion site was found in all VLF chickens but also in 2 LF chickens from Generation 1. The VLF phenotype appears to be associated with OSD because of a founder effect. Indeed, only one Nunukan male was at the origin of the experimental subpopulation studied at INRA, Jouyen-Josas. This male was probably homozygous for OSD but not for the VLF condition. The male, No. 575, which was found in Generation 4 to be homozygous for the VLF condition, was also homozygous for OSD. This male’s crossbred progeny, born from EF dams, were VLF and carried OSD (Figure 4c). This result shows that the VLF phenotype was transmitted in a dominant way by Sire
Consequences of the Deletion at the ev21-K Locus The deletion described in the present study represents a new allele at the ev21 insertion site, which we have named OSD. The 3′ flanking region of OSD seems to be identical to the 3′ flanking region of the OS allele previously described in LF chickens. It seems likely that the deletion in the provirus and the 5′ flanking region occurred after the insertion of a complete provirus. It is not known whether the OSD allele of VLF chickens also carries the 32 kbp deletion described for the OS allele of LF chickens by Iraqi and Smith (1995). Obviously, the retroviral insertion will not be able to lead to the production of a complete viral particle, but could still produce the env protein, as observed with ev6 (Robinson et al., 1981). This point will have to be investigated by a search for RNA expression and the use of an in vitro system that could reveal the presence of the env protein. The important question to address is whether the deletion is responsible for the altered feathering phenotype VLF. The limited number of birds and the sampling strategy used in the present study does not allow for linkage analysis. However, our preliminary information on the inheritance of the VLF phenotype does not support the hypothesis of a sex-linked dominant allele to the K mutation. The structural modification observed at the ev21
FIGURE 6. A PCR test that discriminates between the OS and OSD alleles. A typical 1.5% agarose gel showing the amplified PCR products from reactions A (Lanes 1 to 4) and B (Lanes 5 to 8) of the OSD specific assay. Reaction A contains primers that are specific to the unoccupied endogenous virus (ev) ev21 integration site (US) and the 3′ end of the occupied site (OS). Reaction B discriminates between the OS and OSD alleles. Reference early feathering (EF) chickens carrying the US only yield a band of 1,060 bp in reaction A (Lane 1), and no product in reaction B (Lane 5); late feathering (LF) chickens carrying US+OS haplotype yield the 1,060-bp band (specific to the US) plus a 390-bp band derived from the 3′ end of the OS allele in reaction A (Lane 2), and the 1,115-bp upstream junction fragment in reaction B (Lane 6); very late feathering (VLF) chickens carrying the US+OSD haplotype display an identical pattern to the LF (US+OS) for reaction A (Lane 3), but show a single 600-bp upstream junction fragment in reaction B (Lane 7); EF chickens carrying the OS only yield a single band of 390-bp in reaction A (Lane 4), and the 1,115-bp upstream junction band in reaction B (Lane 8). For details of the PCR assay, see text, Figure 3 and Table 2.
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
PR-GAG and POL-2. Standard EF chickens are free of ev21 and do not show any amplified product in this assay (Figure 6, Lane 5). In chickens carrying the OSD allele, primers PN-UP and POL-2 combine to yield an amplified band of 600 bp (Figure 6, Lane 7). The LF birds (OS+US), and EF birds that carry the OS only, show a band of 1,115 bp due to the interaction of PN-UP and PR-GAG (Figure 6, Lanes 6 and 8). The interaction of PN-UP with POL-2 on DNA from birds carrying the complete ev21 insert would result in a band of over 6 kbp. Fragments of this size do not amplify under standard PCR conditions. This test is fully informative in females but can not determine the haplotypes present in males, owing to the presence of two Z chromosomes.
742
TIXIER-BOICHARD ET AL.
575. Male progeny obtained from VLF breeding females, however, were not always VLF, which invalidates the hypothesis that VLF is determined by one single dominant allele on the Z chromosome. At present, it can be assumed that OSD is required but not sufficient to obtain the VLF phenotype. The PCR diagnostic test developed in the present study is currently being used on a larger number of VLF, LF, and EF chicks from later generations, in order to obtain more data on the relationship between OSD and the VLF phenotype. The crossbred progeny of Sire 575 are being backcrossed to EF females of the WG line, in order to follow the segregation of the feathering phenotype and of OSD.
ACKNOWLEDGMENT The authors are grateful to E. J. Smith, who provided the RAV-2 probe and the ev21-int probe.
Bacon, L. D., E. J. Smith, L. B. Crittenden, and G. B. Havenstein, 1988. Association of the slow-feathering (K) and an endogenous viral (ev21) gene of the Z chromosome of chickens. Poultry Sci. 67:191–197. Benkel, B. F., J. Mucha, and J. S. Gavora, 1992. A new diagnostic method for the detection of endogenous Rousassociated virus-type provirus in chickens. Poultry Sci. 71: 1520–1526. Benkel, B. F., and E. J. Smith, 1993. Research note: A rapid method for the detection of the Rous-associated endogenous solitary long terminal repeat, ev15. Poultry Sci. 72:1601–1605. Boulliou, A., J. P. Le Pennec, G. Hubert, R. Donal, and M. Smiley, 1991. Restriction fragment length polymorphism analysis of endogenous avian leukosis viral loci: determination of frequencies in commercial broiler lines. Poultry Sci. 70:1287–1296. Boulliou, A., J. P. Le Pennec, G. Hubert, R. Donal, and M. Smiley, 1992. The endogenous retroviral ev21 locus in commercial chicken lines and its relationship with the slow-feathering phenotype (K). Poultry Sci. 71:38–46. Chambers, J. R., E. J. Smith, E. A. Dunnington, and P. B. Siegel, 1993. Sex-linked feathering (K,k+) in chickens: a review. Poult. Sci. Rev. 5:97–116. Crittenden, L. B., 1991. Retroviral elements in the genome of the chicken: implications for poultry genetics and breeding. Crit. Rev. Poult. Biol. 3:73–109. Don, R. H., P. T. Cox, B. J. Wainwright, and J. S. Mattick, 1991. “Touchdown” PCR to circumvent spurious priming during gene amplification. Nucl. Acids Res. 19:4008.
Downloaded from http://ps.oxfordjournals.org/ by guest on March 15, 2016
REFERENCES
Gavora, J. S., U. Ku¨hnlein, and J. L. Spencer, 1989. Absence of endogenous viral genes in an inbred line of Leghorn chickens selected for high egg production and Marek’s disease resistance. J. Anim. Breed. Genet. 106:217–224. Higuchi, R., 1989. Simple and rapid preparation of samples for PCR. Pages 31–38 in: PCR Technology: Principles and Applications for DNA Amplifications. H. A. Erlich, ed. Stockton Press, New York, NY. Iraqi, F., and E. J. Smith, 1995. Organization of the sex-linked late-feathering haplotype in chickens. Anim. Genet. 26: 141–146. Levin, I., and E. J. Smith, 1990. Molecular analysis of endogenous virus ev21 slow-feathering complex of chickens. 1. Cloning of proviral-cell junction fragment and unoccupied integrate site. Poultry Sci. 69:2017–2026. Levin, I., and E. J. Smith, 1991. Association of a chicken repetitive element with the endogenous virus-21 slowfeathering locus. Poultry Sci. 70:1948–1956. McGibbon, W. H., 1977. A sex-linked mutation affecting rate of feathering in chickens. Poultry Sci. 56:872–875. Robinson, H. L., S. M. Astrin, A. M. Senior, and F. H. Salazar, 1981. Host susceptibility to endogenous viruses: defective glycoprotein-expressing proviruses interfere with infections. J. Virol. 42:745–751. Ronfort, C., M. Afanassieff, Y. Chebloune, G. Dambrine, V. M. Nigon, and G. Verdier, 1991. Identification and structure analysis of endogenous proviral sequences in a Brown Leghorn chicken strain. Poultry Sci. 70:2161–2175. Schwartz, D. E., R. Tizard, and W. Gilbert, 1983. Nucleotide sequence of Rous sarcoma virus. Cell 32:853–869. Shank, P. R., S. H. Hughes, and H. E. Varmus, 1981. Restriction endonuclease mapping of the DNA of Rous-associated virus-0 reveals extensive homology in structure and sequence with Avian Sarcoma Virus DNA. Virology 108: 177–188. Smith, E. J., and L. B. Crittenden, 1986. Endogenous virus genes in a slow-feathering line of White Leghorns. Avian Pathol. 15:395–406. Smith, E. J., and I. Levin, 1991. Application of a locus-specific DNA hybridization probe in the analysis of the slowfeathering endogenous virus complex of chickens. Poultry Sci. 70:1957–1964. Somes, R. G., Jr., 1969. Delayed feathering, a third allele at the K locus of the domestic fowl. J. Hered. 60:281–286. Tixier-Boichard, M., L. Durand, M. Morisson, F. Ricard, and G. Coquerelle, 1994a. Comparative analysis of avian leukosisrelated endogenous viral genes in experimental strains of the domestic chicken. Genet. Sel. Evol. 26:53s–66s. Tixier-Boichard, M., B. F. Benkel, J. R. Chambers, and J. S. Gavora, 1994b. Screening chickens for endogenous virus ev21 viral element by the polymerase chain reaction. Poultry Sci. 73:1612–1616.