Sequence variants of chicken linker histone H1.a Ewa Go ´ rnicka-Michalska 1 , Jan Pałyga 1 , Andrzej Kowalski 1 and Katarzyna Cywa-Benko 2 1 Department of Genetics, Institute of Biology, Akademia S ´ wie˛ tokrzyska, Kielce, Poland 2 Department of Animal Production, Faculty of Biology and Agriculture, Rzeszo ´ w University, Poland In eukaryotes, the arrays of nucleosomes are linked together by DNA to form chromatin fibers with a diameter of 10 nm. In bulk chromatin, one molecule of histone H1 binds to a linker DNA sealing off two turns of DNA around a core histone octamer. Thus, histone H1 plays a direct role in stabilizing nucleo- somal and higher-order chromatin structures. A histone H1 family in higher eukaryotes is repre- sented by several primary-sequence subtypes (nonallelic variants) encoded by different, albeit closely related genes [1,2]. Disruption of a linker histone gene in both protista and fungi demonstrated [3] that it is not essen- tial for cell survival. On the other hand, elimination of three but not one or two H1 subtypes in mouse leads to embryonic lethality due to about 50% reduction in an H1 to nucleosome ratio [4]. It has been shown that the expression patterns of multiple H1 genes can change during development and tissue differentiation [3]. The particular H1 variants can also be nonrandomly distributed in chromatin, differ in their turnover rates and in the extent of post-translational phosphorylation [5]. Such differences suggest that linker histone variants could play a dis- tinct role in the regulation of chromatin activity. Regu- latory function of the histone H1 may include a selective control of individual gene expression [3,6], inhibition of DNA repair by homologous recombina- tion [7], and transmission of apoptotic signals from the nucleus to the mitochondria [8], as well as targeting and activation of a major apoptotic nuclease DFF40 ⁄ CAD during terminal stages of apoptosis [9]. Heritable developmental defects accompanied by Keywords allelic isoforms; chicken; genetic polymorphism; histone H1; peptide microsequencing Correspondence E. Go ´ rnicka-Michalska, Department of Genetics, Institute of Biology, Akademia S ´ wie˛ tokrzyska, ul. S ´ wie˛ tokrzyska 15; 25-406 Kielce, Poland Fax: +48 41 3496292 Tel: +48 41 3496333 E-mail: egorn@pu.kielce.pl or jpalyga@pu.kielce.pl (Received 22 August 2005, revised 16 January 2006, accepted 19 January 2006) doi:10.1111/j.1742-4658.2006.05147.x Two allelic isoforms (H1.a1 and H1.a2) of histone H1.a were identified within two conservative flocks (R11 and R55) of Rhode Island Red chick- ens. These proteins form three phenotypes: a1, a2 and a1a2. Birds with phenotype a1 were most common (frequency 0.825–0.980) while the a1a2 chickens appeared relatively rarely (0.017–0.175). The third phenotype a2, not detected in the tested populations, has only been revealed in progeny of the purpose-mated a1a2 birds. The polymorphism of histone H1.a was observed in all examined chicken tissues, so that the H1 preparations isola- ted from the lung, spleen, kidney and testis from the same individual exhib- ited identical phenotypes (a1, a2, or a1a2). This finding, together with inheritance data, supports the genetic nature of the H1.a polymorphism. As indicated by cleavages with a-chymotrypsin and protease V8, the H1.a1 and H1.a2 are two highly related proteins which differ within N-terminal part of their C-terminal tails. Only a single nonconservative amino acid substitution between both H1.a allelic isoforms was detected by Edman degradation: glutamic acid present at position 117 in histone H1.a1 was replaced by lysine in histone H1.a2. Furthermore, using microsequencing techniques we have found a sequence homology between the N- and C-ter- minal parts of an unknown minor protein H1.y, present in the phenotype a2, and similar regions of histone H1.b. Abbreviations 2-ME, 2-mercaptoethanol; PMSF, phenylmethylsulfonyl fluoride; PVDF, polyvinylidene difluoride; RBS, 0.01 M Tris ⁄ HCl, pH 7.4, 0.01 M NaCl, 0.003 M MgCl 2 ; NaCl ⁄ Cit, 0.15 M NaCl, 0.015 M sodium citrate; SDS, sodium dodecyl sulfate; TCA, trichloroacetic acid. 1240 FEBS Journal 273 (2006) 1240–1250 ª 2006 FEBS changes in DNA methylation following a knockdown of histone H1 genes in Arabidopsis [10], as well as dis- turbances in chromatin structure leading to reduced DNA methylation at specific regulatory sites and ⁄ or altered expression of specific genes in triple-H1 null mouse embryonic stem cells [11] suggest that linker histones could operate as epigenetic regulators of higher-order chromatin conformations. Several studies support the idea that individual H1 subtypes could exhibit functional specialization. For example, histone B4, a Xenopus embryonic linker his- tone, may confer a totipotency to embryonic cells [12] while H1b, a specific subtype of mouse histone H1 complement, can restrain muscle differentiation by forming a complex with a homeoprotein Msx1 [13]. It has been demonstrated [14] that H1.1, one out of seven linker histones in Caenorhabditis elegans, took part in chromatin silencing in a germ line. In addition, this protein was also able to enhance a telomeric position effect in budding yeast [15]. The authors suggested that specific structural features of the H1.1 variant could dedicate this protein to epigenetic silencing. Metazoan histone H1 proteins have a characteristic tripartite structure which consists of a central globular domain flanked by two basic N- and C-terminal tails that are largely unstructured in a physiological solu- tion [1]. However, these disordered regions can assume a secondary structure when interacting with DNA [16,17]. While the globular domain is the most highly conserved region amongst somatic histone H1 variants, the sequences of H1 tails are more variable, both in length and amino acid composition, indicating that they evolved by insertion or deletion and nucleotide substitutions [18]. By expression of domain-swapping mutants, Brown et al. [6] determined that the central globular domains of H1 0 and H1c were responsible for their differential effects on gene expression. Recently, using targeted mutations, Hendzel et al. [19] showed that the C-ter- minal domain is required for a high affinity binding of the histone H1 to chromatin. Moreover, they discov- ered differences in the binding affinity of the individual histone H1 subtypes in vivo attributed to the subtype- specific properties of the C-terminus [20]. Several nonallelic H1 variants may also be represen- ted by two or more allelic isoforms in the chromatin of several species [21–25] including humans [26]. Dif- ferences in the amino acid sequences between allelic variants of a given H1 subtype were demonstrated in only a few instances. They were located mainly in ter- minal domains of H1 molecule and usually originated as a result of amino acid substitutions [25–28] or inser- tions and deletions of repeated sequence motifs [21,24]. As in the case of the nonallelic variants [20], it seems that the alterations in molecular structure of allelic H1 isoforms may affect their affinity to DNA and chroma- tin, and their ability to interact with other chromatin proteins. Most probably, this would result in certain phenotypic effects [24,29,30]. In our laboratory we have revealed a polymorphism for subtypes H1.a, H1.b, and H1.z in duck [23,31–33] and quail [34,35]. In this work, in addition to well- known allelic variants of erythrocyte histone H5 [27] distributed differently in distinct chicken breeds [36], we also observed two allelic isoforms of histone H1.a in the chickens with a rare histone H1.a2 in a Rhode Island Red breed only. Herein, we have demonstrated the genetic nature of the chicken histone H1.a poly- morphism and have employed a limited enzymatic cleavage and amino acid microsequencing to find a difference in the primary structure of the H1.a1 and H1.a2 allelic isoforms. Results Variability of histone H1.a in Rhode Island Red chickens Histone H1 proteins were extracted with a diluted per- chloric acid solution from erythrocyte nuclei of indi- vidual chickens and analyzed in an acetic acid–urea polyacrylamide gel. As seen in Fig. 1, the perchloric acid-soluble proteins predominantly contained a fast moving histone H5 and six slow migrating H1 sub- types. They were designated according to the nomencla- ture of [37] as H1.a,.a¢,.b,.c,.c¢ and .d. Two erythrocyte isoforms of histone H1.a (H1.11 L), H1.a1 and H1.a2, were identified in the acetic acid–urea gel within Rhode Island Red chicken populations. These variants form three phenotypes: a1, a2 and a1a2 (Fig. 1). The animals with phenotype a1 possess a high level of a single pro- tein H1.a1. Histone H1.a from the birds with phenotype a1a2 was resolved into two well-separated bands, H1.a1 and H1.a2, with a slightly lower level of the H1.a2 isoform. The third phenotype a2 has only been found following purposely mating the birds with phenotype a1a2. In the a2 individuals, in addition to a high level of the expected histone H1.a2, a low amount of either H1.a1 or H1.a1-like protein (H1.y) was discerned (Fig. 1). The differences in the electrophoretic mobility between H1.a1 and H1.a2 in the acetic acid–urea gel (Figs 1 and 2) and a lack of such differences in the poly- acrylamide gel electrophoresis conducted in the presence of SDS (Fig. 3) seem to indicate that both proteins have a similar apparent molecular weight and can differ from each other in a net charge. E. Go ´ rnicka-Michalska et al. Variants of chicken linker histone H1.a FEBS Journal 273 (2006) 1240–1250 ª 2006 FEBS 1241 A phenotypic variability of histone H1.a was detec- ted only in the Rhode Island Red chickens (Table 1). In the R11 and R55 conservative flocks, we found two phenotypes, a1 and a1a2. Birds with phenotype a1 were most common (frequency 0.825–0.980) while the heterozygous a1a2 chickens were quite rare (frequency 0.017–0.175). Moreover, the frequency of the pheno- type a1a2 was very similar in two different genetic groups (R11 and R55) and varied widely between two independent populations of the same conservative flock R11 (Table 1). A distribution of the phenotypes of erythrocyte his- tone H1.a in a progeny from various matings of the Rhode Island Red chickens is presented in Table 2. As we were unable to detect animals with phenotype a2 in all tested farm populations, we could only perform three available types of crosses: a1 · a1, a1a2 · a1a2, and a1 · a1a2. As expected, the third phenotype a2 was revealed after mating the a1a2 individuals. In the offspring of such parents, we found three categories of phenotypes: a1, a1a2 and a2 in a ratio 1 : 2 : 1. The data obtained from all crosses support the assumption that a single gene with two co-dominant alleles is responsible for the observed polymorphism of histone H1.a. The histone H1 preparations isolated from lung, spleen, kidney and testis of three Rhode Island Red Fig. 1. A comparison of three phenotypes of erythrocyte histone H1.a in Rhode Island Red chickens. Histone H1 proteins from erythrocytes nuclei were resolved in acetic acid–urea polyacryl- amide gel. H1 and H5, histone H1 and H5; H1.a,.a¢,.b,.c,.c¢,.d, his- tone H1subtypes; H1.a1, H1.a2, two isoforms of histone H1.a; H1.y, H1.a1-like protein. The phenotypes of histone H1.a were des- ignated as a1, a2, a1a2. Fig. 2. A comparison of histone H1.a from erythrocytes (E), lung (L), spleen (S), kidney (K) and testis (T) of chicken individuals with phenotype a1, a2 and a1a2. Proteins were separated in acetic acid– urea polyacrylamide gel. All symbols are explained in the legend to Fig. 1. Fig. 3. Concentration and purification of histone H1.a isoforms and H1.y protein from chicken erythrocyte nuclei by a preparative SDS– gel electrophoresis. The bands containing H1.a and H1.y were excised from the gel, avoiding contamination with remnants of H1.a¢. Variants of chicken linker histone H1.a E. Go ´ rnicka-Michalska et al. 1242 FEBS Journal 273 (2006) 1240–1250 ª 2006 FEBS chicken phenotypes were also resolved in the acetic acid-urea gel (Fig. 2) into six main subtypes [37], which exhibited tissue specificity (Fig. 2). A poly- morphism of histone H1.a was observed in all exam- ined chicken tissues (Fig. 2) so that H1 isolated from the lung, spleen, kidney and testis of the same individ- ual exhibited identical phenotypes (a1, a2 or a1a2). These results, together with inheritance data (Table 2), support the genetic nature of the H1.a polymorphism. Limited proteolysis of H1.a isoforms and protein H1.y The stained protein bands containing appropriate all- elic forms of the chicken erythrocyte histone H1.a or protein H1.y were cut out from the acetic acid–urea gel. After concentration and purification by prepara- tive sodium dodecyl sulfate (SDS) electrophoresis (Fig. 3), the proteins were recovered from gel strips by electroelution and then precipitated from the solution. Samples of the H1.a and H1.y prepared in such a way were suitable for further analysis of the protein struc- ture. By using a limited digestion with a-chymotrypsin, we revealed that the polymorphic isoforms of histone H1.a differed in their C-terminal peptides (Fig. 4A). As his- tone H1 molecule is cleaved with a-chymotrypsin on the C-terminal side of a single phenylalanine [32], a limited a-chymotrypsin digestion of the chicken histone H1.a yielded two fragments [37]: a C-peptide that rep- resented the C-terminal part of the H1.a (residue 110–224) and a slightly shorter N-peptide from the N-terminus to Phe109 (Fig. 5). In the acetic acid–urea gel, a relative migration of a-chymotrypsin generated C 1 - and C 2 -peptides resembled that of the native pro- teins H1.a1 and H1.a2 (Fig. 4A), while that of the N-peptides was identical regardless of the H1.a iso- form. Moreover, molecular masses of the chymotryptic N-terminal peptides from H1.a1 and H1.a2 isoforms determined by an electrospray ionization mass spectro- metry were found to be identical except for a hetero- geneity within the H1.a1 N-peptide that appeared to be represented by a mixture of N-acetylated and N-un- acetylated species (results not shown). Collectively, these results clearly indicate that a putative difference in the amino acid sequence between H1.a isoforms is located in the region spanning amino acid residue 110 (arginine) and the C-terminus (Fig. 5). Similarly, one of the two peptides obtained by H1.y digestion with a-chymotrypsin migrated in the gel like H1.a1- or H1.a2-derived N-peptide, whereas the second one, with an electrophoretic mobility similar to H1.a C-peptides, was almost undetectable in the gel (Fig. 4A). To demonstrate which part of the C-terminal domain is responsible for putative amino acid differences between both isoforms, the H1.a1 and H1.a2 were Table 1. A frequency of phenotypes and alleles of erythrocyte histone H1.a in conservative flocks of four chicken breeds. Chickens from the same genetic group, R11, were bred in two different farms in Szczytno near De˛blin and Chorzelo ´ w near Mielec. Breed and genetic group Total number of birds Number of birds with phenotypes Frequency of phenotypes Frequency of alleles a1 a1a2 a2 a1 a1a2 a2 a1 a2 Rhode Island Red R55 50 49 1 – 0.980 0.020 0 0.99 0.01 R11 (Szczytno) 60 59 1 – 0.983 0.017 0 0.99 0.01 R11 (Chorzelo ´ w) 126 104 22 – 0.825 0.175 0 0.91 0.09 Leghorn G99 157 157 – – 1 0 0 1 0 Cornish CE2 61 61 – – 1 0 0 1 0 Greenleg Partridge Z11 52 52 – – 1 0 0 1 0 Table 2. Distribution of phenotypes of erythrocyte histone H1.a in progeny from various matings of Rhode Island Red R11 chickens. Type of mating Number of Frequency of phenotypes in progeny (observed ⁄ expected) v 2 Male Female Families Progeny a1 a1a2 a2 a1 a1 6 25 25 ⁄ 25 a1a2 a1a2 2 67 15 ⁄ 16.75 36 ⁄ 33.5 16 ⁄ 16.75 0.404 a1a2 a1 2 54 26 ⁄ 27 28 ⁄ 27 0.074 E. Go ´ rnicka-Michalska et al. Variants of chicken linker histone H1.a FEBS Journal 273 (2006) 1240–1250 ª 2006 FEBS 1243 digested with Staphylococcus aureus protease V8. After histone H1.a treatment with the protease V8 that can preferentially cleave a peptide bond following glutamic acid, only one distinct peptide band, identified as C-ter- minal tail of H1.a (results not shown), was observed in the gel (Fig. 4B). In the C-terminal domain of H1.a molecule there are only two glutamic acid residues at positions 117 and 120 [37]. Thus, a limited digestion of H1.a with protease V8 generated a little shorter C-ter- minal peptide ranging from residue 118 or 121 to the C-terminus (Fig. 5). Moreover, the remaining part of the histone H1.a, with several glutamic acid residues (positions 2, 10, 46, 57 and 78), would have to undergo further cleavages into a number of smaller fragments [37] undetectable in the gel. As seen in Fig. 4B, the elec- trophoretic mobilities of the C-terminal peptides obtained from H1.a1 and H1.a2 isoforms following digestion with protease V8 were similar. As we revealed distinct differences in a migration between C 1 - and C 2 -peptides derived from H1.a1 and H1.a2 proteins after limited proteolysis with a-chymotrypsin, the uni- form migration of protease V8-cleaved C-peptides seemed to point out that differences between H1.a iso- forms were confined to a short region spanning from amino acid residue 110 to the glutamic acid at either the position 117 or 120 (Fig. 5). As mentioned for H1.a isoforms, a protease V8 digestion of the protein H1.y also yielded only one peptide (Fig. 4B). Tandem mass spectrometry To find more precisely which site could be altered between the two allelic H1.a isoforms, we applied a tandem mass spectrometry for a sequence prediction of the peptides generated by trypsin digestion. As a sim- ilar set of peptides was detected in both proteins (results not shown) we were unable to reveal any meaningful alterations between the isoforms using this procedure. However, a primary structure of protein H1.y was partially revealed by the tandem mass spectrometric analysis of its trypsin products. Besides five pep- tides present in almost all of the H1 subtypes, the trypsin digestion generated one peculiar peptide (AETAPVAAPDVAAAPTPAK), which showed a high degree of similarity to histone H1.b (H1.03) (resi- dues 1–19). This 19-amino acid peptide was a part of Fig. 4. Limited proteolysis of H1.a1 and H1.a2 histones, and H1.y (H1.a1-like protein) from chicken erythrocytes with a-chymotrypsin (A) and Staphylococcus aureus V8 protease (B). The digestion prod- ucts were resolved in the acetic acid–urea gel. C 1 ,C 2 , chymotryptic C-terminal peptides generated from the H1.a1 and H1.a2, respect- ively; N, chymotryptic N-terminal peptides produced from the H1.a1, H1.a2 and H1.y; C, V8 protease C-terminal peptides obtained from the H1.a1, H1.a2 and H1.y; H1, undigested histone H1 subtypes. Fig. 5. The main peptides generated by a limited cleavage of the chicken erythrocyte histone H1.a isoforms with a-chymotrypsin and Staphylococcus aureus protease V8. H1.a1 and H1.a2, isoforms of histone H1.a; C and N, C- and N-termini of histone H1.a; (C 1 ,C 2 )- chym., chymotryptic C 1 -andC 2 -peptides generated from the H1.a isoforms; N-chym., chymotryptic N-peptides produced from the H1.a1 and H1.a2; C-V8, the possible protease V8 C-peptides cleaved from the H1.a. Using a published sequence [37], we assumed that limited proteolysis of H1.a with the protease V8 cre- ated a C-terminal peptide running from residue 118 or 121 to the C-terminus. The expected difference between H1.a isoforms is located at a short region spanning from Arg110 to the glutamic acid either at position 117 or 120. Variants of chicken linker histone H1.a E. Go ´ rnicka-Michalska et al. 1244 FEBS Journal 273 (2006) 1240–1250 ª 2006 FEBS the N-terminal domain, the most variable and specific region for each H1 subtype [37]. As the individual ion score for H1.y N-peptide was 75 and this value was significantly higher than the threshold score of 45, it strongly supports either the identity or very extensive homology (P<0.05) between this peptide and the his- tone H1.b (data not shown). Partial microsequencing of chymotryptic C-terminal fragments We also applied automated Edman degradation for a partial sequence determination in the H1.a1 and H1.a2 proteins in the region affected by a mutation. For this purpose, the C-terminal fragments of histone H1.a iso- forms produced by a-chymotrypsin digestion have been analyzed by amino acid sequencing after trans- blotting the proteins from the SDS–polyacrylamide gel to a polyvinylidene difluoride (PVDF) membrane. Fig- ure 6 shows a partial sequence comparison of chicken histone H1.a variants. In a region spanning the resi- dues 110–120 in both H1.a1 and H1.a2 we revealed 10 identical amino acids out of 11 residues so that the all- elic isoforms differed only by one amino acid. A gluta- mic acid at the position 117 in the histone H1.a1 was replaced by a lysine in the H1.a2 isoform. The partial histone H1.a2 sequence data was deposited in the Uni- Prot Knowledgebase (UniProtKB, http://www.ebi. ac.uk) under the accession number P8451533. We also determined the first 11 amino acid residues for the longer C-terminal chymotryptic peptide cleaved off the chicken H1.y (Fig. 6). The analyzed fragment corresponded exactly to the beginning of the C-ter- minal domain of the histone H1.b (residues 109–119; with serine at position 6 and aspartic acid at position 8 of the peptide) and that of the histone H1.a2 (with proline at position 6 and lysine at position 8 of the peptide). As the protein H1.y was isolated from the chicken phenotype a2 enriched in the isoform H1.a2, the bands used for a histone H1.y concentration were cut out from the acetic acid-urea gel in which H1.y and H1.a2 migrated close to each other (Fig. 1). Further separation of H1.y protein from the histone H1.a2 could not be achieved by the preparative SDS–gel elec- trophoresis due to almost identical migration of both proteins (Fig. 3). Therefore, we conclude that our his- tone H1.y preparation was sequenced as a mixture of the histone H1.y and an accompanying histone H1.a2. Thus, the automated Edman degradation of H1.y chymotryptic C-peptide revealed a sequence homology between the beginning of the C-terminal part of H1.y and that of H1.b (residues 109–119; Fig. 6). Discussion Each of the six H1 genes (11L, 11R, 03, 0.10, 01 and 02) identified in the chicken genome [38] encodes for a different H1 subtype [37]. Differences in amino acid sequences among H1 subtypes could be connected with their differential capacity for interaction with chroma- tin [1,20]. In a current dynamic view [39], the H1 mole- cules, with a residence time on the nucleosome modulated by a phosphorylation, are continuously exchanged between chromatin binding sites in a ‘stop- and-go’ mode. A specificity in amino acid sequence may also preferentially alter the chromatin residence time of a unique variant at a specific locus or chroma- tin region [20,39]. Thus, a microheterogeneity in the primary structure may imply functional differences between distinct H1 subtypes. The linker histone het- erogeneity can further increase due to a genetic poly- morphism and ⁄ or post-translational modifications [36,40,41]. In this study, two allelic isoforms (H1.a1 and H1.a2) of histone H1.a were identified within two conservative flocks (R11 and R55) of Rhode Island Red chickens. Histone H1.a also exhibits a distinct intrapopulation variability in other avian species such as duck [31] and quail [34]. As in the chicken, the poly- morphic H1.a subtype from duck erythrocytes was rep- resented by two electromorphs, H1.a1 and H1.a2, with similar apparent molecular weights and different net charges [31]. The genetic polymorphism of histone H1.a in Pharaoh quail was associated with a lack of this proteins in some birds or its occurrence at either an intermediate or high level in other individuals [34]. In both flocks of Rhode Island Red chickens (Table 1), representing a typical multipurpose breed with a brownish red plumage, a frequency of allele a2 was quite low (0.01–0.09). Similarly, the allele a2, which was also rare or absent in most genetic duck groups, was enriched in the color-feathered duck popu- Fig. 6. Alignment of the first 11 amino acid residues in the chymo- tryptic C-peptides obtained from chicken histones H1.a1, H1.a2 and protein H1.y. Chicken histone H1.a (residues 110–120) and H1.b (residues 109–119) sequences designated with the asterisk [Uni- Prot Knowledgebase (UniProtKB) accession numbers P08287 and P08285, respectively] are presented for comparison. E. Go ´ rnicka-Michalska et al. Variants of chicken linker histone H1.a FEBS Journal 273 (2006) 1240–1250 ª 2006 FEBS 1245 lations [31]. Surprisingly, the number of homozygous a2 chickens was not reduced and the proportion of other phenotypes did not differ from the expected val- ues in the offspring of purpose-mated heterozygous birds. As the homozygous individuals a2 did not seem to be less viable, it is likely that a lack of this pheno- type in both conservative chicken flocks might be either fortuitous or caused by breeding conditions. The changes in histone H1 allele frequencies were observed in quail populations divergently selected for a high or low reduction in body mass following transient starva- tion [30]. Only a1a1 homozygotes and a1a0 hetero- zygotes were detected among individuals of those populations, while rare homozygotes a0a0 were found exclusively in a control random mating population [30]. Thus, it was likely that a divergent selection in quail could have acted against the allele a 0 . Berdnikov et al. [42] analyzed the influence of environment on the histone H1 allele frequencies in Pisum sativum. Using a collection of 833 accessions of the cultivated pea ori- ginating from different regions of the Old World they revealed that the alleles of the specific H1 subtypes (H1–5 and, possibly, H1–1) were subjected to climatic- ally dependent natural selection under conditions of primitive farming. The similar molecular weights and different net charges of the chicken H1.a1 and H1.a2 suggest that one of these proteins might have arisen as a result of amino acid substitution or post-translational modifica- tion. In order to check whether phosphorylation is responsible for H1.a heterogeneity, we performed a digestion of H1 samples from the individuals differing in the H1.a phenotype with alkaline phosphatase (results not shown). As the enzyme treatment did not change the patterns of histone H1.a migration, we believed that phosphorylation is not responsible for the observed polymorphism. Thus, the most convin- cing explanation of the intrapopulation variability of H1.a is that the H1.a1 and H1.a2 represent allelic pro- teins encoded by a gene with two codominant alleles at a locus. Previously, we detected [36] two allelic electro- morphs (a and b) of the chicken erythrocyte histone H5. A limited proteolysis with a-chymotrypsin and subsequent microsequencing (unpublished results) revealed that both polymorphic variants differed in their N-terminal tails and therefore they resembled two chicken H5 sequence variants Va and Vb with a gluta- mine-to-arginine substitution at the position 15 [27]. Most of the allelic variants so far reported for the polymorphic histone H1 subtypes differ in the N- or C-terminal tails [23,24,27,32], though an amino acid replacement in the globular domain was also detected [25]. Specific site cleavages in H1.a1 and H1.a2 pro- teins with a-chymotrypsin and protease V8 have shown that a variable region of the H1.a isoforms is located at the beginning of the C-terminal tail, between amino acid residue 110 and glutamic acid either at position 117 or 120. Relying on the sequence data for the chicken histone H1 subtypes [37] and their genes [38], a limited amino acid variability is apparent in the part of C-terminal tail adjacent to the globular domain. In this region there are several conservative (N ⁄ S, E ⁄ D, V ⁄ G) and nonconservative (V⁄ T, P⁄ S, K ⁄ L) substitutions between paralogous H1 subtypes including those involving charged residues. The C-terminal domain of histone H1 is presumably involved in the organization of the linker DNA and may be responsible for stabilization of condensed chro- matin fiber [19,43]. Vila et al. [16] have studied con- formational properties of a C-terminal peptide CH-1, which is placed adjacent to the central globular domain in rodent histone H1 0 (residues 99–121), by a Fourier- transform infrared spectroscopy. The authors showed the presence of inducible helical and b-turn (S ⁄ TPKK) elements in this peptide in both trifluoroethanol solu- tion and in the complexes with DNA. In the presence of helical inducers, the helical region has a marked am- phipatic nature with all basic residues on one site of the helix and all hydrophobic residues on the other one [17]. The amino acid substitutions in this H1 region may influence its binding to chromatin [19], including binding to specific chromatin proteins, for histone H1 C-terminal region may mediate protein–protein interac- tions [9,44]. Lu and Hansen [43] demonstrated that the ability of the H1 0 to alter linker DNA conformation and to stabilize condensed chromatin structure is not distributed evenly throughout the entire domain. The sequences that mediate folding were localized to two specific subdomains within C-terminal tail (97–121 and 146–169). Interestingly, the two H1 0 S ⁄ TPKK motifs were found in each of the C-terminal subdomain (resi- dues 97–121 and 146–169). These sequences form b-turns that bind the minor groove of DNA [45]. A comparison of sequence data of the C-terminal region responsible for H1.a variability indicates that allelic isoforms of the chicken histone H1.a differ by only one amino acid residue. At position 117, there is a glutamic acid in histone H1.a1 and lysine in histone H1.a2. At the nucleotide level, a transition from G to A in the first position of glutamic acid codons (GAA or GAG) might have resulted in the replacement of Glu117 to Lys (codons: AAA or AAG) in H1.a2. Thus, a charge difference between both H1.a allelic isoforms was due to a nonconservative Glu117Lys substitution which could in turn explain a different mobility of H1.a1 and H1.a2 in the acetic acid–urea gel. Variants of chicken linker histone H1.a E. Go ´ rnicka-Michalska et al. 1246 FEBS Journal 273 (2006) 1240–1250 ª 2006 FEBS The amino acid sequence variation among different paralogous H1 subtypes might influence a positive charge of the protein and thereby affect the interaction between negatively charged DNA and positively charged C-terminal tails. Recent studies have shown [46,47] that a modulation of the coulombic interactions between H1 and DNA by post-translational modifica- tions or by amino-acid mutations is essential for gene regulation. Thus, the substitution of a negatively charged amino acid (Glu) in the chicken H1.a1 to a positively charged one (Lys) in the H1.a2 is an essen- tial sequence alteration with possible functional conse- quences. The electrostatic DNA–histone interactions are supplemented by the regulated interactions between histones and effector proteins in the chromatin region posed for gene transcription, as proposed by the ‘histone code’ hypothesis [48]. Histone covalent modifi- cations, especially on core histone N-tails, act sequen- tially or in a combination to form a nucleosomal ‘epigenetic code’ that can differentially interact with structural, enzymatic and ⁄ or regulatory protein com- plexes [48]. As suggested [10], the members of the his- tone H1 family which could also harbor epigenetic markings [40] may act themselves as epigenetic regula- tors. Lately, some experiments seem to provide evi- dence for functional specificity among H1 subtypes [8,12,13,44]. Recently, Berdnikov et al. [24] compared quantita- tive traits in near isogenic lines of lentil and grasspea carrying allelic variants of the most abundant subtype H1-1. In this experiment, small (1–8%) but significant differences have been revealed for some quantitative traits such as: the mean number of seeds per pod, mean seed mass, the flower fresh mass and the number of ovules in a carpel. These effects might have reflected a direct association between histone H1 complement and fitness, as evidenced for a Lys178Asn substitution in the C-terminus of H1–5)1, a slow allelic variant of pea histone H1–5 [25] that was most abundant in geo- graphical region with a cold climate [42]. At present, the functional significance of allelic vari- ation within the relevant H1 subtype is unknown. Although no satisfactory assay for assessing their role in chromatin is currently available [26], we [30] and others [42] observed a differential susceptibility of H1 allelic variants to artificial and natural selection, respectively, in response to environmental stimuli that most likely might challenge a genetic buffering capacity of Hsp90 chaperone complexes and ⁄ or other capacitors of phenotypic variation [49]. Recent observation [50] that Hsp90 facilitates loading linker histones to a speci- fic binding and transporting protein NASP seems to support a putative role of the chaperone in buffering phenotypic effects of genetic variation in H1 histones under normal conditions. Experimental procedures Animals In this study we used Rhode Island Red chickens main- tained as the conservative flocks (R11 and R55) at the breeding poultry farm in Szczytno near De˛blin and in Chorzelo ´ w (R11) near Mielec, Poland. The Cornish chick- ens were obtained from Poultry Research and Development Centre in Zakrzewo near Poznan ´ . All the remaining chicken breeds were from the farm in Szczytno near De˛blin. Blood from wing vein of individual birds was separately collected into NaCl ⁄ Cit solution (0.15 m NaCl, 0.015 m sodium citrate) containing 0.1 mm phenylmethylsulfonyl fluoride (PMSF). Further experiments were carried out using lung, spleen, kidney and testis from three 5-month-old indi- viduals of the Rhode Island Red R11 breed. After quick decapitation, the tissues were immediately removed, perfused or washed with ice-cold NaCl ⁄ Cit solution supplemented with phenylmethylsulfonyl fluoride and stored at )20 °C, but no longer than three weeks. These experiments were car- ried out in accordance with the European Communities Council Directive of November 24 1986 (86/609/EEC). Isolation of erythrocytes, nuclei and histone H1 extraction All procedures were performed at 0–4 °C. The blood mixed with NaCl ⁄ Cit solution was centrifuged at 2000 g for 10 min (MPW-360, MPW Med. Instruments, Warsaw, Poland). The supernatant with a layer of leukocytes was aspirated, and the erythrocytes were washed twice with NaCl ⁄ Cit solution containing 0.5 mm phenylmethylsulfonyl fluoride. The pellet of red blood cells was stored at )20 °C until used. Nuclei were isolated by one of two methods: (a) Erythro- cyte nuclei were prepared by a lysis in 0.1 m phosphate buffer, pH 7.0, containing 0.9% (w ⁄ v) NaCl, 0.03% (w ⁄ v) saponin and 1 mm phenylmethylsulfonyl fluoride [28]. (b) Nuclei from lung, spleen, kidney and testis were pre- pared by a sucrose-Triton X-100 method [51]. Histone H1 proteins were extracted from the washed nuclei first with 1 m and then with 0.5 m perchloric acid. Protein from the combined supernatants were precipitated with 20% (w ⁄ v) trichloroacetic acid, washed twice with aci- dified acetone (acetone ⁄ concentrated HCl, 250 : 1, v ⁄ v) and acetone, and then air dried. Electrophoretic analysis Histone H1 proteins were separated in an acetic acid–urea gel containing 15% (w ⁄ v) acrylamide, 0.9 m acetic acid and E. Go ´ rnicka-Michalska et al. Variants of chicken linker histone H1.a FEBS Journal 273 (2006) 1240–1250 ª 2006 FEBS 1247 8 m urea and stained in Coomassie Blue R-250 [32]. For preparative SDS gel electrophoresis, approximately 18 stained gel fragments of the acetic acid–urea gel containing H1.a isoforms and H1.a1-like protein (H1.y) were sepa- rately incubated in adaptation buffer [2% (w ⁄ v) SDS, 10% (v ⁄ v) 2-mercaptoethanol (2-ME), 10% (v ⁄ v) glycerol, 0.125 m Tris ⁄ HCl, pH 6.8]. The gel pieces were then stacked in a well of one-dimensional 15% (w ⁄ v) polyacryl- amide gel containing 0.1% (w ⁄ v) SDS [35]. Isolation of histone H1.a isoforms and protein H1.y The stained bands of the appropriate protein, purified and concentrated during preparative electrophoresis, were excised from the SDS gel to avoid neighboring contamina- tions [35]. Three protein bands of the same kind (H1.a1, H1.a2 or H1.y) were pooled into glass tubes, and put over a dialysis membrane. Electroelution was performed in a buffer containing 0.1% (w ⁄ v) SDS, 0.025 m Tris, and 0.192 m glycine. Protein samples precipitated with 0.2 m KCl were centrifuged at 10 000 g for 10 min (K24D, Janetzki, Leipzig, Germany). Precipitated SDS was then removed by washing the pellets with 20% (w ⁄ v) trichloroacetic acid. A small amount of pellet remaining after centrifugation was washed twice with acidified acetone and acetone, and dried. Digestion with a-chymotrypsin and protease V8 from Staphylococcus aureus Histone H1.a isoforms and protein H1.y isolated by elec- troelution were dissolved in: (a) 10 lL 0.05 m Tris ⁄ HCl, pH 8.0 and a-chymotrypsin (45 UÆmg )1 , Serva Electrophor- esis, Heidelberg, Germany) was added to a concentration of 0.2 ngÆlL )1 ; (b) 10 lL 0.1 m Tris ⁄ HCl, pH 7.5 and pro- tease V8 (660 UÆmg )1 , Sigma, St. Louis, MO, USA) was added to final enzyme concentration 7 ngÆlL )1 . In both cases, the proteolysis was performed at room temperature for 15 and 45 min, respectively. The reaction was stopped by boiling the samples for 2 min. Then appropriate volumes of 8 m urea and concentrated acetic acid, 2-ME and gly- cerol were added to receive 2.5 m urea, 0.9 m acetic acid, 5% (v ⁄ v) 2-ME and 10% (v ⁄ v) glycerol in the reaction solution. The proteolytic degradation products were resolved in the acetic acid–urea polyacrylamide gel. Identification of protein resolved in the gel by mass spectrometry The strips of the acetic acid–urea gel containing H1.a iso- forms and protein H1.y, stained with Coomassie blue R-250, were equilibrated with SDS-adaptation buffer. Only one gel piece per well was loaded into the wells of 15% (w ⁄ v) polyacrylamide gel containing SDS. After electro- phoresis, the protein bands were excised, in-gel reduced, S-alkylated and digested with modified trypsin (Promega, Madison, WI, USA). Tryptic peptides, washed out from the gels, were resolved by RP-HPLC using LC Packings (Sunnyvale, CA, USA) C 18 precolumn (300 lm · 5 mm) and C 18 column (75 lm · 15 cm). Peptides were eluted at 0.2 lLÆmin )1 with a linear gradient of acetonitrile from 0% to 25% (v ⁄ v) in 25 min in the presence of 0.05% (v ⁄ v) for- mic acid. The column outlet was coupled to an electrospray mass spectrometer (Micromass, Manchester, UK). Molecular mass analysis was performed using the nano- Z-spray ion source of quadrupole-time of flight electrospray mass spectrometer working with the regime of data depend- ent on MS to tandem MS switch, allowing for a 3-s sequen- cing scan for each detected peptide. The data were analyzed using the mascot program (http://www.matrixscience.com). All electrospray experiments were carried out at the mass spectrometry facility of the Institute of Biochemistry and Biophysics of the Polish Academy of Sciences in Warsaw, Poland. In-gel digestion, electroblotting and amino acid analysis Stained bands containing concentrated and purified iso- forms of histone H1.a and protein H1.y were cut out from the SDS-preparative gel and soaked in adaptation buffer [0.1% (w ⁄ v) SDS, 1 mm EDTA, 10% (v ⁄ v) glycerol, 0.125 m Tris ⁄ HCl, pH 6.8]. Next, the samples were loaded into the separate wells of 15% (w ⁄ v) polyacrylamide gel containing 0.1% (w ⁄ v) SDS and overlayered with 4 lLof the adaptation buffer containing 1 lgÆlL )1 of a-chymotryp- sin (45 UÆmg )1 , Serva Electrophoresis). After electrophor- esis, the proteins and polypeptides were electroblotted for 1.5 h at 90 V onto PVDF membrane (Roche, Mannheim, Germany) using 0.01 m 3-cyclohexylamino-1-propane- sulfonic acid, 10% (v ⁄ v) methanol, pH 11 as a transfer buf- fer. The PVDF membrane was washed in MilliQ (Millipore, Bedford, MA, USA) deionized water, stained with 0.1% (w ⁄ v) Coomassie Blue R-250 in 40% (v ⁄ v) methanol and 1% (v ⁄ v) acetic acid. After destaining in 50% (v ⁄ v) meth- anol, the filter was rinsed in MilliQ deionized water and air dried. Finally, the bands of chymotryptic C-peptides derived from H1.a isoforms and protein H1.y were excised from the blots and analyzed by automated Edman sequencing. The N-terminal amino acid sequence analysis was per- formed on a gas-phase sequencer (Model 491, Perkin Elmer-Applied Biosystems, Foster City, CA, USA) at Bio- Center (Jagiellonian University, Krako ´ w, Poland). The phenylthiohydantoin derivatives were analyzed by on-line gradient high performance liquid chromatography on Microgradient Delivery System Model 140C equipped with Programmable Absorbance Detector Model 785 A (both from Perkin Elmer-Applied Biosystems). Variants of chicken linker histone H1.a E. Go ´ rnicka-Michalska et al. 1248 FEBS Journal 273 (2006) 1240–1250 ª 2006 FEBS Acknowledgements This work was supported by grant KBN 6P06D03520 from the Polish Ministry of National Education and Science. References 1 Cole RD (1987) Microheterogeneity in H1 histones and its consequences. Int J Pept Protein Res 30, 433–449. 2 Sullivan S, Sink DW, Trout KL, Makalowska I, Taylor PM, Baxevanis AD & Landsman D (2002) The histone database. Nucleic Acids Res 30, 341–342. 3 Khochbin S (2001) Histone H1 diversity: bridging regu- latory signals to linker histone function. Gene 271, 1–12. 4 Fan Y, Nikitina T, Morin-Kensicki EM, Zhao J, Ma- gnuson TR, Woodcock CL & Skoultchi AI (2003) H1 linker histones are essential for mouse development and affect nucleosome spacing in vivo . Mol Cell Biol 23, 4559–4572. 5 Parseghian MH & Hamkalo BA (2001) A compendium of the histone H1 family of somatic subtypes: an elusive cast of characters and their characteristics. Biochem Cell Biol 79, 289–304. 6 Brown DT, Gunjan A, Alexander BT & Sittman DB (1997) Differential effect of H1 variant overproduction on gene expression is due to differences in the central globular domain. Nucleic Acids Res 25, 5003–5009. 7 Downs JA, Kosmidou E, Morgan A & Jackson SP (2003) Suppression of homologous recombination by the Saccharomyces cerevisiae linker histone. Mol Cell 11, 1685–1692. 8 Konishi A, Shimizu S, Hirota J, Takao T, Fan Y, Mats- uoka Y, Zhang L, Yoneda Y, Fujii Y, Skoultchi AI et al. (2003) Involvement of histone H1.2 in apoptosis induced by DNA double-strand breaks. Cell 114, 673– 688. 9 Widlak P, Kalinowska M, Parseghian MH, Lu X, Hansen JC & Garrard WT (2005) The histone H1 C-terminal domain binds to the apoptotic nuclease, DNA Fragmentation Factor (DFF40 ⁄ CAD) and stimu- lates DNA cleavage. Biochemistry 44, 7871–7878. 10 Wierzbicki AT & Jerzmanowski A (2005) Suppression of histone H1 genes in Arabidopsis results in heritable developmental defects and stochastic changes in DNA methylation. Genetics 169, 997–1008. 11 Fan Y, Nikitina T, Zhao J, Fleury TJ, Bhattacharyya R, Bouhassira EE, Stein A, Woodcock CL & Skoultchi AI (2005) Histone H1 depletion in mammals alters glo- bal chromatin structure but causes specific changes in gene regulation. Cell 123, 1199–1212. 12 Saeki H, Ohsumi K, Aihara H, Ito T, Hirose S, Ura K & Kaneda Y (2005) Linker histone variants control chromatin dynamics during early embryogenesis. Proc Natl Acad Sci USA 102, 5697–5702. 13 Lee H, Habas R & Abate-Shen C (2004) MSX1 coop- erates with histone H1b for inhibition of transcription and myogenesis. Science 304, 1675–1678. 14 Jedrusik MA & Schulze E (2001) A single histone H1 isoform (H1.1) is essential for chromatin silencing and germline development in Caenorhabditis elegans. Devel- opment 128, 1069–1080. 15 Jedrusik MA & Schulze E (2003) Telomeric position effect variegation in Saccharomyces cerevisiae by Cae- norhabditis elegans linker histones suggests a mechanistic connection between germ line and telomeric silencing. Mol Cell Biol 23, 3681–3691. 16 Vila R, Ponte I, Collado M, Arrondo JL & Suau P (2001) Induction of secondary structure in a COOH- terminal peptide of histone H1 by interaction with the DNA: an infrared spectroscopy study. J Biol Chem 276, 30898–30903. 17 Vila R, Ponte I, Jimenez MA, Rico M & Suau P (2000) A helix-turn motif in the C-terminal domain of histone H1. Protein Sci 9, 627–636. 18 Ponte I, Vila R & Suau P (2003) Sequence complexity of histone H1 subtypes. Mol Biol Evol 20, 371–380. 19 Hendzel MJ, Lever MA, Crawford E & Th’ng JP (2004) The C-terminal domain is the primary determinant of histone H1 binding to chromatin in vivo. J Biol Chem 279, 20028–20034. 20 Th’ng JP, Sung R, Ye M & Hendzel MJ (2005) H1 family histones in the nucleus: control of binding and localization by the C-terminal domain. J Biol Chem 280, 27809–27814. 21 Gorel FL, Berdnikov VA & Rozov SM (1982) Mole- cular variants of histone H5 of linnet (Acanthis flam- mea). Mol Biol (Mosk) 16, 790–798 (in Russian). 22 Zweidler A (1984) Core histone variants of the mouse: Primary structure and differential expression. In Histone Genes: Structure, Organization and Regulation (Stein, GS Stein, JL & Marzluff, WF, eds), pp. 339–371. John Wiley, New York. 23 Pałyga J, Go ´ rnicka-Michalska E, Kowalski A & Ksia˛ _ zki- ewicz J (2000) Natural allelic variation of duck erythro- cyte histone H1b. Int J Biochem Cell Biol 32, 665–675. 24 Berdnikov VA, Bogdanova VS, Gorel FL, Kosterin OE & Trusov YA (2003) Large changes in the structure of the major histone H1 subtype result in small effects on quantitative traits in legumes. Genetica 119, 167–182. 25 Bogdanova VS, Lester DR, Berdnikov VA & Andersson I (2005) Structure of allelic variants of subtype 5 of his- tone H1 in pea Pisum sativum L. Heredity 94, 582–588. 26 Sarg B, Green A, Soderkvist P, Helliger W, Rundquist I & Lindner HH (2005) Characterization of sequence var- iations in human histone H1.2 and H1.4 subtypes. FEBS J 272, 3673–3683. 27 Greenaway PJ & Murray K (1971) Heterogeneity and polymorphism in chicken erythrocyte histone fraction V. Nat New Biol 229, 233–238. E. Go ´ rnicka-Michalska et al. Variants of chicken linker histone H1.a FEBS Journal 273 (2006) 1240–1250 ª 2006 FEBS 1249 [...].. .Variants of chicken linker histone H1.a 28 Neelin JM, Neelin EM, Lindsay DW, Pałyga J, Nichols CR & Cheng KM (1995) The occurrence of a mutant dimerizable histone H5 in Japanese quail erythrocytes Genome 38, 982–990 29 Bogdanova VS, Rozov SM, Trusov YA & Berdnikov VA (1994) Phenotypic effect of substitutions of short chromosomal segments containing different alleles of histone H1 genes... (1998) Isolation and preliminary characterization of histone H1.b allelic variants from quail erythrocytes Genome 41, 709–719 ´ ´ 36 Gornicka-Michalska E, Pałyga J, Lubon H, Kowalski A & Cywa-Benko K (1998) Genetic variants of chicken erythrocyte histone H5 Biochem Mol Biol Int 44, 605– 615 37 Shannon MF & Wells JR (1987) Characterization of the six chicken histone H1 proteins and alignment with their respective... Characterization of the chicken histone H1 gene complement: generation of a complete set of vertebrate H1 protein sequences J Biol Chem 262, 9656–9663 39 Bustin M, Catez F & Lim JH (2005) The dynamics of histone H1 function in chromatin Mol Cell 17, 617– 620 1250 ´ E Gornicka-Michalska et al 40 Kuzmichev A, Jenuwein T, Tempst P & Reinberg D (2004) Different Ezh2-containing complexes target methylation of histone. .. frequencies formed in the course of Pisum sativum L (pea) cultivation Heredity 71, 199–209 43 Lu X & Hansen JC (2004) Identification of specific functional subdomains within the linker histone H1o C-terminal domain J Biol Chem 279, 8701–8707 44 Montes de Oca R, Lee KK & Wilson KL (2005) Binding of barrier-to-autointegration factor (BAF) to histone H3 and selected linker histones including H1.1 J Biol Chem... sativum L.) Genet Res 64, 35–41 30 Pałyga J (1998) Distribution of allelic forms of erythrocyte H1 histones in Japanese quail populations divergently selected for amount of weight loss after transient starvation Biochem Genet 36, 79–92 ´ 31 Kowalski A, Pałyga J, Gornicka-Michalska E & Krajewska WM (1998) Allelic polymorphism of histone H1.a in duck erythrocytes Biochem Genet 36, 183–191 ´ 32 Pałyga... (2001) Translating the histone code Science 293, 1074–1080 49 Sangster TA, Lindquist S & Queitsch C (2004) Under cover: causes, effects and implications of Hsp90mediated genetic capacitance Bioessays 26, 348–362 50 Alekseev OM, Widgren EE, Richardson RT & O’Rand MG (2005) Association of NASP with HSP90 in mouse spermatogenic cells: stimulation of ATPase activity and transport of linker histones into nuclei... (1989) SPKK, a new nucleic acid-binding unit of protein found in histone EMBO J 8, 797–804 46 Dou Y & Gorovsky MA (2000) Phosphorylation of linker histone H1 regulates gene expression in vivo by creating a charge patch Mol Cell 6, 225–231 47 Dou Y, Bowen J, Liu Y & Gorovsky MA (2002) Phosphorylation and an ATP-dependent process increase the dynamic exchange of H1 in chromatin J Cell Biol 158, 1161–1170... histone H1 or nucleosomal histone H3 Mol Cell 14, 183–193 41 Sarg B, Helliger W, Talasz H, Forg B & Lindner HH (2005) Histone H1 phosphorylation occurs site-specifically during interphase and mitosis: identification of a novel phosphorylation site on histone H1 J Biol Chem, doi: 10.1074/jbc.M508957200 42 Berdnikov VA, Bogdanova VS, Rozov SM & Kosterin OE (1993) Geographic patterns of histone H1 allelic frequencies... Gornicka-Michalska E & Kowalski A (1993) Genetic polymorphism of histone H1.z in duck erythrocytes Biochem J 294, 859–863 ´ 33 Kowalski A, Pałyga J & Gornicka-Michalska E (2004) Identification of histone H1.z components in a Muscovy duck (Cairina moschata L.) population Comp Biochem Physiol B Biochem Mol Biol 137, 151–157 34 Pałyga J (1998) Genes for polymorphic H1 histones are linked in the Japanese quail genome... NASP with HSP90 in mouse spermatogenic cells: stimulation of ATPase activity and transport of linker histones into nuclei J Biol Chem 280, 2904–2911 51 Busch H & Daskal Y (1977) Methods for isolation of nuclei and nucleoli Methods Cell Biol 15, 1–43 FEBS Journal 273 (2006) 1240–1250 ª 2006 FEBS . Sequence variants of chicken linker histone H1. a Ewa Go ´ rnicka-Michalska 1 , Jan Pałyga 1 , Andrzej Kowalski 1 and Katarzyna Cywa-Benko 2 1 Department. that our his- tone H1. y preparation was sequenced as a mixture of the histone H1. y and an accompanying histone H1. a2 . Thus, the automated Edman degradation