a large number of lysine and arginine residues binds to linker DNA between ..... of the chicken H5, in most polymorphic histone H1 variants usually two alleles at ...
Vol. 44, No. 3, March 1998 BIOCHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL Pages605-615
G E N E T I C V A R I A N T S O F C H I C K E N E R Y T H R O C Y T E H I S T O N E H5 Ewa G6rnicka-Michalska 1'* Jan Patyga ~, Henryk Lubofi 1, Andrzej Kowalski ~, Katarzyna CywaBenko 2 ~Department of Genetics, Wyzsza Szkota Pedagogiczna, 15 Konopnickiej Street, '25-406 Kielce, Poland 2Institute of Animal Production, 32-083 Balice near Krak6w, Poland Received November 6, 1997 S U M M A R Y : T w o allelic electromorphs a and b of chicken eryhrocyte histone H5 have been detected in a sodium dodecyl sulfate polyacrylamide gel. In an acid-urea gel, however, each of the allelic variants was found to be accompanied by a slower migrating form. A comparison of c~-chymotrypsin-digested products of H5.a and H5.b revealed that they differed in N-terminal domains. The H5 variants were distributed differently not only in various chicken races but also in distinct lines within a breed. Allele H5 b was about 2.6-4.6 as abundant as its counterpart H5" in most chicken populations examined. These proportions were distorted in two Leghorn lines: the ratio of H5 b to H5" was only 1.6 in line H22 and increased up to 32 in line G99. Key words: Chicken, erythrocyte, histone H5, polymorphism INTRODUCTION Silencing of avian e~ythrocyte chromatin is caused by maintaining a high level of histone H5 [1] which is also able to condense and inactivate chromatin when expressed ectopically in nonerythroid cells [2, 3]. Histone H5 consists of a globular central domain flanked by N- and C-terminal tails. The globular domain is involved in nucleosome binding, whereas the C-terminal tail containing a large number of lysine and arginine residues binds to linker DNA between nucleosomes and is likely to contribute to folding of nucleosomes into higher order chromatin filament [4]. The role o f the N-terminal tail has not
yet been precisely defined. It is, however, a site for
polymorphism of histone H5 in the chicken. Two sequence variants: Va and Vb differing in glutamine-to-arginine substitution at position 15 were identified by Amberlite CG-50 column
Abbreviations used: EDTA, ethylene diamine tetraacetic acid disodium salt; NBS, N-bromosuccinimidc: PMSF, phenylmethylsulfonyl fluoride; SDS, sodimn dodecylsulfate; SSC, 0.15 M NaCI, 0.015 M sodium citrate; TEMED, N,N,N' N'-tetramethylethylene diamine. *
1 To whom correspondence should be addressed 1039-9712/98/030605- [ l $05.00/0 605
Copyright 9 1998 by Academic Press Australia. All rights of reproduction in any form reserved.
Vol. 44, No. 3, 1998
BIOCHEMISTRYond MOLECULARBIOLOGYINTERNATIONAL
chromatography [-5]. The Glu/Arg polymorphism of the chicken H5 was subsequently confirmed by amino acid sequencing [6]. These two H5 variants were also detected using high pressure liquid chromatography [7] and capillary electrophoresis [8]. Recently we have discovered [9] an unusual polymorphism o f erythrot~te historic l~5 in a population descended from a feral Japanese quail from Hawaii. One of the allelic ~a, a,,,s contained a suithydryl group as evidenced by a spomaneous formation of protein dimers in the polyacrylamide gel in the absence of reducing agents and by reactions with sulfhydryl-spec~fic reagents. This polymorphism was not detected in other quail swains tested. In this work we compared a distribution o f allelic forms o f histone H5 in various chicken breeds. As we have found that chicken variants of this histone could be distinguished in sodium dodecyl sulfate~polyacrylamide gel, we employed this electrophoretic technique for their population screening. We have also provided evidence for genetic basis of histone H5 polymorphism in the chicken and preliminary data for identification o f affected domain in our H5 histone preparations. MATERIALS AND M E T H O D S
Animals Six chicken breeds were used: Cu-eenleg Partridge, Yellowleg Partridge, Sussex, Rhode Island Red, Leghorn and Cornish. The Cornish chickens were obtained from the Poultry Research and Development Centre in Zakrzewo near Poznafi, Poland. All the remaining chicken breeds were from the farm of genetic reserve flocks in Szczytno near D~blin, Poland. Blood from individual chickens was collected into centrifuge tubes filled with SSC* solution containing 0.1 mM CdSO4 and 0.1 mM PMSF to a 1/3 of their height. The blood samples were transported to the laboratory on ice. Isolation of the erythrocyte nuclear proteins soluble m 0.5 M HCI04 solulion Red blood cells were separated from a diluted serum by centrifugation at 2000 g at 4~ The buffy coat was removed by aspiration and after resuspending the blood cells were centrifuged twice more in SSC solution containing 0.1 mM CdSO4 and 0.1 mM PMSF. Erythrocyte nuclei were isolated by a lysis in 0.9% NaCi containing 0.03% saponin and 1 mM PMSF, followed by repeated washes in saline-PMSF [9]. Lysine-rich histones (H 1 and H5) were obtained by the extraction of the washed erythrocyte nuclei first with 1 M and then with 0.5 M perchloric acid. The combined acid extracts were precipitated with 20% trichloroacetic acid and the proteins were washed twice with acidified acetone (acetone:conc. HCI, 250:1) and acetone, then air-dried. All procedures were performed at 0 - 4~ Electrophoretic mlalysis Proteins were analyzed either in a 15% polyacrylamide slab gel containing 0.1% SDS or in an acetic acid-urea slab gel containing 15% acrylamide, 0.9 M acetic acid and 8 M urea [10]. Cleavage of historic H5 with N-bromosuccinimide The Coomassie Brilliant Blue R250-stained bands of the histone H5 (phenotypes a, b and ab) were cut out of the acid-urea gel and each gel strip was placed separately into a freshly prepared solution containing 1 mg NBS in 50% acetic acid. After 30-min incubation in the
606
Vol. 44. No. 3. 1998
BIOCHEMISTRYond MOLECULAR BIOLOGY INTERNATIONAL
dark, a second portion (1 mg) of a freshly prepared NBS solution was added. Following an additional 30-rain incubation in the presence of NBS, the gel fragments with cleaved H5 were transfered to new tubes and incubated for 3x15 min in 8 M urea, 0.9 M acetic acid, "% TEMED and 10% glycerol, and then placed on top of acetic acid-urea polyao~ylamide slab gel. For SDS-polyacrylamide gel electrophoresis the gel pieces containing the cleaved H5 were equilibrated for 3x15 min in an adaptation buffer (0.1% SDS, 10% glycerol, 0 125 M Tris/HCl, pH 6.8 and 1 mM EDTA), and then the gel pieces were applied directly on the surface of the SDS-polyacrylamide slab gel. Cleavage o f histone H5 with ct-chymotrypsm The SDS gel fragments containing histone H5 (phenotypes a, b and ab) were first soaked for 3x15 min in the SDS adaptation buffer (see above), and then the equilibrated gel strips were incubated for 30 vain in 0.3 ml of the adaptation buffer containing 200 lag of otchymotrypsin (Serva, Heidelberg, Germany). Proteolysis was stopped by boiling the samples for 2 min. The gel pieces prepared in this manner were placed on top of SDS-polyacrylamide slab gel. Alternatively, the strips with cleaved histone H5, after inactivation of the enzyme, were also preincubated for 3x15 min in the solution containing 8 M urea, 0.9 M acetic acid, 2% TEMED and 10% glycerol, and then placed on top of acid-urea polyacrylamide slab gel. RESULTS AND DISCUSSION
As shown in Fig. 1, perchloric acid-soluble proteins from chicken erythrocyte nuclei were resolved both in sodium dodecyl sulfate and acid-urea polyacrylamide gels into H1 and H5 histones. It was found that histone H5 isolated from some chicken individuals could migrate in the SDS gel as a single protein band while from others as a double band. As the single H5 protein from the erythrocytes ofhomozygous birds migrated either as a fast (H5a) or slow (H5b) band, the heterozygous birds contained both aUelic proteins. A more complicated electrophoretic pattern emerged when chicken histone H5 was separated in the acid-urea polyacrylamide gel since each of the homozygous H5 proteins which had migrated as a single band in the SDS gel was resolved in the acid-urea gel into two separate protein bands (H5al and H5a2 or H5bl and H5b2). The H5 histone from the heterozygous birds was separated into three bands among which the middle broad one was the most prominent, probably due to the overlapping of the band HSal with the band H5b2 (Fig. 1.I). The source of the heterogeneity o f the H5 bands from homozygous individuals in the acid-urea gel remains unclear. It is well known [11] that phosphorylation of histone H5 in mature erythrocytes is very low. In accordance with earlier observations [5], a digestion of our H5 histone preparations with alkaline phosphatase did not change the patterns of this protein in the acid-urea gel (results not shown). There is, however, conflicting evidence regarding a degree of H5 histone acetylation. While Ruiz-Carillo et al. [12] reported incorporation of radioactive acetate into duck H5 which steadily diminished as erytrocytes maturated, Sung et al. [13] have found that chicken erythrocyte histone H5 was neither acetylated nor methylated. Furthermore, Greenaway and
607
BIOCHEMISTRY(mdMOLECULAR
Vol. 4 4 , No. 3, 1 9 9 8 (I)
BIOLOGY INTERNATIONAL
(II) a
b
ab
a
b
ab
H,I .51-
-- H5b -- H5a H5a2 -H 5 a l --
Fig. 1. Phenotypes of chicken erythrocyte histone H5 in sodium dodecyl sulfate (I) and acetic acid-urea (II) polyacrylamide gel; H1 and H5, histones H1 and H5, respectively. In the SDS gel (I) histone H5 migrated either as a single band (phenotypes a and b containing only subfraction H5a and H5b, respectively) or as a doublet (phenotype ab with both subfractions, i.e. H5a and H5b). In the acid-urea gel (II) the histone H5 from homozygous individuals was resolved into two protein bands (phenotypes a and b) while that from heterozygous chickens migrated as a triple band (phenotype ab). A protein doublet H5bl and H5b2 from homozygotes bb migrated slightly faster then that for proteins H5al and H5a2 from homozygotes aa.
Murray [5] concluded that N-terminal threonine in chicken H5 was not acetylated but S .ligy ei
al. [14] noticed that the recovery of the N-threonyl residue from H5 histones in five avian species was lower than expected for an unblocked residue. We cannot also exclude that hi~tt,ne H5 doublets in acid-urea gel may arise as a result of other modifications of histone 115. I~ r example, it has been found [15] that histone H5 could be poly-(ADP)-ribosylated in vitro. The inheritance data for the phenotypes of the histone H5 in the chicken (Table I) appears to indicate that this protein is coded by a gene with two codominant alleles, tt5" and
H5 b, at a locus. The distribution of the three phenotypes of erythrocyte histone H5 in the populations of six chicken breeds is shown in Table 2. As the homozygotes with phenotype b predominated in all chicken races, with the exception of Leghorn H22, the homozygotes with phenotype a were relatively rare (0 - 18%). The distribution of heterozygotes (phenotype ab) ranged from 21% in Cornish to 44% in Sussex with only 5.8% in Leghorn G99. On the overall, the allele H5 b, encoding the subfraction H5b, was found to be a main form in the all breeds examined with the
608
Vol. 44, No. 3, 1998
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL
Table 1. Distribution of phenotypes of erythrocyte histone H5 in progeny from variou chl~ke'l matings Type of mating
Number of
male x female families aa ab 1 aa bb 1 bb ab 7 bb bb 49
progeny 2 5 35 151
Distribution ofphenotypes (observ./expect.) a 1/1
ab 1/1 5/5 17/17.5
b
18/17.5 151/151
0 0.029 ~ 0
*For this backcross mating there is no significant difference between observed and expected distribution of H5 phenotypes.
frequency ranging from 0.61 to 0.97. The frequency of the H5 alleles in Leghorn differed significantly from that in other breeds, i.e. primitive lines (Yellowleg and Greenleg), meat chicken (Cornish) and the two other egg-laying breeds (Sussex and Rhode Island Red). Moreover, H5 allele levels also differed significantly between the two Leghorn lines. To establish which part of the chicken H5 molecule is responsible for the observed polymorphism, the protein bands from each of histone phenotypes (a, b and ab) were cut out of the acid-urea gel and treated with N-bromosuccinimide or ot-chymotrypsin which can specifically cleave the polypeptide bonds in |ysine-rich histones at the tyrosine and phenylalanine residues, respectively. After digestion of histone H5 with NBS (Fig. 2) only C-terminal polypeptides resulting from the cleavage at tyrosine residues 28, 53 and 58 were observed both in the SDS and acidurea gels in which the fastest-migrating band C3 was the most prominent. N-terminal peptides were not detected because their low molecular weights probably hampered their identification by Coomassie staining. The gel patterns of C-peptides generated by NBS cleavage were virtually similar in the all chicken H5 phenotypes what seemed to suggest that N-terminal tails of the allelic forms were likely to be affected. A variable gel pattern of H5 was revealed (Fig. 3) following digestion of each phenotype with ct-chymotrypsin. Aside from the C-terminal peptides of roughly the same apparent molecular weights, the N-peptides with distinct mobilities in the SDS gel were detected (Fig. 3.I). An N,-peptide derived from the histone H5 from birds with phenotype a migrated faster than Nb-peptide from the chickens with phenotype b. Heterozygous individuals possessed both kinds of N-peptides: N~, and Ni, (Fig. 3.1). lu the SDS gel, the migration of the N-peptides generated by digestion with cz-chymoirypsin
609
Total number of birds 50 50 50 60 49 223 61 ,,, 6.0 4.0 4.0 6.7 18.4 0 11.5
a
ab 26.0 28.0 44.0 43.3 40.8 5.8 21.3
b 68.0 68.0 52.0 50.0 40.8 94.2 67.2
Frequency of phenotype (in %)
a b c d e abcdef cdef
0.19 0.18 0.26 0.28 0.39 0.03 0.22
115"
0.81 0.82 0.74 0.72 0.61 0.97 0.78
H5 b
Significant Frequency of allele differencesw
u v w x uvwxyz uvwxyz z
Significant differences~
~Significant differences (•2, p
