Whole blood samples for histone H1 isolation were collected separately from 34 Guinea fowl individuals belonging to separate white- and black-feathered subpopulations bred at the Department of Poultry Breeding of the University of Technology and Life Sciences in Bydgoszcz, Poland.

The blood samples collected into SSC solution (0.15 M NaCl and 0.015 M sodium citrate) containing 1 mM phenylmethylsulfonyl fluoride (PMSF) as a protease inhibitor were centrifuged at a low speed and then erythrocyte pellets were washed twice with the same solution. The purified nuclei were isolated from the washed erythrocyte pellet by using 3% saponin in 0.1 M sodium phosphate buffer, pH 7.0, supplemented with 0.9% NaCl and 1 mM PMSF [27]. The lysed nuclei were washed three more times with 0.9% NaCl in the above buffer containing 1 mM PMSF.

Acid-soluble proteins were extracted from the purified erythrocyte nuclei by a gentle homogenization with 1 M perchloric acid solution. The pellet obtained after centrifugation was subsequently homogenized with the solution containing 0.5 M perchloric acid. The proteins were precipitated from the pooled supernatants with 20% trichloroacetic acid and air-dried after washing with acetone acidified with HCl (500:2, v/v) followed by pure acetone. Protein preparations were dissolved in a sample buffer (1 mg/200 μL) containing 8 M urea, 0.9 M acetic acid and 10% 2-mercaptoethanol (2-ME).

Histone H1 preparations were analyzed in a two-dimensional polyacrylamide gel. In a first dimension, the proteins were resolved in an acetic acid-urea polyacrylamide gel (15% acrylamide, 0.5% N,N’-methylenebisacrylamide, 8 M urea, 0.9 M acetic acid). The Coomassie Blue R-250 stained stripes of histone H1 were cut out from the gel and incubated twice in an adaptation buffer (2.1% SDS, 2% 2-ME, 10% glycerol, 0.124 M Tris-HCl, pH 6.8) for 15 min. The equilibrated stripes were then electrophoresed in a second dimension in an SDS-polyacrylamide slab gel (13.5% acrylamide, 0.36% N′,N′-methylenebisacrylamide, 0.1% SDS).

To obtain a sufficient amount of pure protein for chemical cleavage and enzymatic digestion, on average, ten bands of the appropriate histone H1.b or H1.c alloform which had been resolved in the acetic acid-urea polyacrylamide gel were separately collected and incubated in the adaptation buffer (see above). Then, the equilibrated gel pieces were electrophoresed in a preparative SDS polyacrylamide gel to concentrate the protein and remove any contamination [28].

The excised bands with concentrated histone H1.b or H1.c alloforms were soaked twice with 1 mL solution of 0.1% N-bromosuccimide (NBS) in 50% acetic acid. After 30 min incubation, the reaction was repeated with freshly prepared NBS solution. Prior to electrophoresis in one dimension 13.5% SDS-polyacrylamide gel, the gel stripes were equilibrated (2 × 15 min) in the adaptation buffer (see above).

Enzymatic digestion of histone H1.b or H1.c alloforms with α-chymotrypsin or S. aureus protease V8 was conducted directly in a stacking SDS-polyacrylamide gel. In both cases, the bands with concentrated protein excised from the preparative gel were soaked (2 × 15 min) in the incubation buffer consisting of 0.1% SDS, 10% glycerol, 0.125 M Tris-HCl, pH 6.8 and 1 mM EDTA. Then, the equilibrated gel pieces were put inside a well in the SDS stacking polyacrylamide gel and overlaid with 2 μL aliquot of the appropriate enzyme dissolved in the incubation buffer (conc. 0.5 mg/500 μL).

Two-dimensional gel images were captured by Doc-Print II gel documentation system (Vilber Lourmat) and processed by using ImageJ 1.42q software (www. rsbweb.nih.gov/ij). The volume of each protein spot was measured as a level of grey in the two-dimensional gel images from three independent individuals. The volume of protein spots was expressed as a percentage of H1.b or H1.c spot(s) relative to all histone H1 spots in the preparation. The coefficient of variation (CV) was calculated as a ratio of standard deviation and mean value. The ratio of protein amount in double spots in heterozygotes was assessed by measuring the volume of each protein constituent within a freehand selected boundary.

Histone H1.b and H1.c were identified along with remaining histone H1 subtypes (H1.a, H1.c’, H1.d and H1.z) in the gel patterns of perchloric acid-soluble protein fraction extracted from Guinea fowl erythrocyte nuclei. In a first dimension acetic acid-urea polyacrylamide gel, the adjacent histone H1.b and histone H1.c bands migrated with similar electrophoretic mobilities in all preparations screened (Fig. 1). More complex patterns of histone H1.b and histone H1.c migration were apparent in a second dimension SDS-polyacrylamide gel (Fig. 2A). A relatively stable migration of histone H1.c′ and H1.d was employed to distinguish between slow (H1.b1 and H1.c1) and fast migrating (H1.b2 and H1.c2) alloforms of the H1.b and H1.c subtypes, respectively. While slow b1 and c1 spots migrated with similar mobilities as those of H1.c′, the faster spots b2 and c2 migrated closer to histone H1.d (Fig. 2A). Differential electrophoretic mobilities of histone H1.b and H1.c alloforms both in the two dimension polyacrylamide gel (Fig. 2A) and one dimension SDS-gel (Fig. 2B) appear to indicate that slower moving H1.b1 and H1.c1 alloforms may differ from faster migrating H1.b2 and H1.c2 counterparts in apparent molecular weights. These alloforms were shown to combine (Fig. 2A) to form three genotypes of histone H1.b (b1b1, b2b2 and b1b2) and histone H1.c (c1c1, c2c2 and c1c2). The phenotypes b1, b2, c1 and c2 contain a single protein, H1.b1, H1.b2, H1.c1 and H1.c2, respectively, while the phenotypes b1b2 and c1c2 are composed of both H1.b1 and H1.b2 or H1.c1 and H1c2 variants. The CV calculated for the histone H1.b and histone H1.c protein spots in all Guinea fowl phenotypes was below a threshold of 0.25 (Table 1), indicating a low variability in the relative spot density between the protein preparations. The volumes of histone H1b and H1c spots in heterozygous individuals were measured in the gel patterns of total histone H1 from three independent individuals and expressed as a mean of grey level. The following results were obtained: 103.82 ± 1.55 for b1 and 103.41 ± 1.85 for b2 in the individuals b1b2 and 104.52 ± 0.38 and 102.77 ± 0.72 for c1 and c2 in the individuals c1c2. The ratio of the spots in the heterozygotes approaching the value 1:1 demonstrates a codominant nature of the constituent alleles at a locus.

The results presented in Tables 2 and 3 show a differential structure of Guinea fowl populations. While the black-feathered population conformed to Hardy-Weinberg proportions (histone H1.b χ² = 1.96, p = 0.16 and histone H1.c χ² = 0.64, p = 0.72), the genotype frequency of white-feathered birds deviated considerably from expected Hardy-Weinberg equilibrium for both loci (χ² = 18.0, d.f = 1, p << 0.001). The inbreeding coefficient (F) equals 0.354 and 0.20 for histones H1.b and H1.c, respectively, in black-feathered subpopulation and, as expected, equals 1 for both H1 loci in white-feathered birds, indicating a possible operation of inbreeding in the tested Guinea fowl group. Thus, the genetic diversity considering the total Guinea fowl group indicated a departure from Hardy-Weinberg proportions (F_IS = 0.610 and 0.634 for H1.b and H1.c loci, respectively) between tested subpopulations as well as in the entire population (F_IT = 0.637). The differences in allele distribution between both strains probed using the Fisher's exact test were statistically significant (p < 0.001 for H1.b and p = 0.002 for H1.c).

To obtain polypeptide maps for identification of a region in H1.b and H1.c molecules responsible for differences between the alloforms, the appropriate protein bands from homozygous individuals were separately concentrated in a preparative 13.5% SDS-gel after a prior resolution in the acetic acid-urea polyacrylamide gel. Histone H1 alloforms were freed from a contaminating histone H1.c′, then in-gel treated with NBS, α-chymotrypsin or S. aureus protease V8 to cleave the histone H1 molecule into specific peptides.

The NBS cleavage of histone H1.c alloforms at a single Tyr residue (Figs. 3 and 4) resulted only in C-terminal peptides in the SDS-polyacrylamide gel because low molecular weight N-terminal peptides could not be detected by staining with Coomassie Blue. As the C-terminal peptide derived from either H1.b1 or H1.c1 alloform migrated slower than its counterpart released from the H1.b2 or H1.c2 alloform, the expected differences in amino acid sequences between tested alloforms might be located in the C-terminal region.

To establish the location of a variable region in the histone H1.b and H1.c alloforms, we have also used α-chymotrypsin and S. aureus protease V8. It appeared (Fig. 3) that digestion with both enzymes provided shorter variable C-terminal peptides that slightly differed in a size and longer invariable N-terminal peptides. As shown in Fig. 3, the C-terminal peptides released by α-chymotrypsin digestion of the histone H1.b1 and H1.c1 alloforms migrated faster in the SDS-polyacrylamide gel than their counterparts from histones H1.b2 and H1.c2. As the chymotrypsin cleaves the peptide bond at the carboxylic side of a single Phe residue, the allele-specific regions of Guinea fowl histone H1.b and H1.c alloforms may have been localized between Arg107 residue and the very C-terminal amino acid (Fig. 4A). A similar pattern of C-peptide migration was also seen (Fig. 3) after digestion of the histone H1 alloforms with S. aureus protease V8. This enzyme can cleave the peptide bonds at Asp or Glu residues, especially at acidic amino acid residues occurring at the beginning of C-terminal domain. As seen in Fig. 3, the protease V8 generated a single slower moving C-peptide released by a presumable digestion at the Glu residue in the C-terminal domain and two faster migrating N-peptides, one of which was likely to be produced by an additional cleavage at the acidic amino acid residue located either in N- or C- terminal domain. A cleavage at the remaining Glu residues in the globular domain would result in much smaller N-peptides that probably would not be visible in the gel. The similar rate of the migration of N-terminal peptides and slightly different mobilities of C-terminal peptides in the SDS-polyacrylamide gel (Fig. 3) following protease V8 digestion of the H1.b and H1.c alloforms appears to indicate that a possible amino acid alteration may be located in the region spanning the onset of C-terminal domain and the very C-terminal end of the molecule.

In general, the above data on the peptide mapping appear to demonstrate that a difference in the amino acid composition in the Guinea fowl histone H1.b and H1.c alloforms may be placed between the NBS-cleaved bond and the C-terminal Lys residue, that is in the C-ending part of H1.c globular domain and/or within the whole C-terminal one (Fig. 4B).