Characterization of sequence variations in human histone H1.2 and H1.4 subtypes ´ Bettina Sarg1, Anna Green2, Peter Soderkvist2, Wilfried Helliger1, Ingemar Rundquist2 ă and Herbert H Lindner1 Division of Clinical Biochemistry, Biocenter, Innsbruck Medical University, Austria Division of Cell Biology, Linkopings Universitet, Sweden ă Keywords HILIC; linker histones; sequence variants; SNP; tumor cell lines Correspondence H H Lindner, Division of Clinical Biochemistry, Biocenter, Innsbruck Medical University, Fritz-Pregl-Strasse 3, A-6020 Innsbruck, Austria Fax: +43 512 507 2876 Tel: +43 512 507 3521 E-mail: herbert.lindner@uibk.ac.at (Received 10 March 2005, revised 10 May 2005, accepted 26 May 2005) doi:10.1111/j.1742-4658.2005.04793.x In humans, eight types of histone H1 exist (H1.1–H1.5, H1°, H1t and H1oo), all consisting of a highly conserved globular domain and less conserved N- and C-terminal tails Although the precise functions of these isoforms are not yet understood, and H1 subtypes have been found to be dispensable for mammalian development, it is now clear that specific functions may be assigned to certain individual H1 subtypes Moreover, microsequence variations within the isoforms, such as polymorphisms or mutations, may have biological significance because of the high degree of sequence conservation of these proteins This study used a hydrophilic interaction liquid chromatographic method to detect sequence variants within the subtypes Two deviations from wild-type H1 sequences were found In K562 erythroleukemic cells, alanine at position 17 in H1.2 was replaced by valine, and, in Raji B lymphoblastoid cells, lysine at position 173 in H1.4 was replaced by arginine We confirmed these findings by DNA sequencing of the corresponding gene segments In K562 cells, a homozygous GCCfiGTC shift was found at codon 18, giving rise to H1.2 Ala17Val because the initial methionine is removed in H1 histones Raji cells showed a heterozygous AAAfiAGA codon change at position 174 in H1.4, corresponding to the Lys173Arg substitution The allele frequency of these sequence variants in a normal Swedish population was found to be 6.8% for the H1.2 GCCfiGTC shift, indicating that this is a relatively frequent polymorphism The AAAfiAGA codon change in H1.4 was detected only in Raji cells and was not present in a normal population or in six other cell lines derived from individuals suffering from Burkitt’s lymphoma The significance of these sequence variants is unclear, but increasing evidence indicates that minor sequence variations in linker histones may change their binding characteristics, influence chromatin remodeling, and specifically affect important cellular functions The H1 histones are small basic proteins occurring in all higher eukaryotes in multiple subtypes that differ only slightly in their primary sequences H1 histones consist of a central, highly conserved globular domain, while the hydrophilic N- and C-terminal tails exhibit less sequence conservation In addition to the heterogeneity of their primary structures, the H1 tails are also extensively post-translationally modified (e.g phosphorylated or ADP-ribosylated) under various biological conditions Moreover, the proportion of H1 Abbreviations CE, capillary electrophoresis; HILIC, hydrophilic interaction liquid chromatography; HPCE, high performance capillary electrophoresis; RFLP, restriction fragment length polymorphism; SNP, single nucleotide polymorphism; TEAP, triethylammonium phosphate FEBS Journal 272 (2005) 3673–3683 ª 2005 FEBS 3673 H1 microsequence variants in human cell lines subtypes varies in a tissue- and species-specific manner, and the expression of each subtype varies throughout development and differentiation [1–3] In the human genome, genes encoding eight different subtypes of histone H1 have been identified The genes encoding H1.1, H1.2, H1.3, H1.4, H1.5 and H1.t are located on the short arm of chromosome 6, while H1° is located on chromosome 22 [4] and H1oo on chromosome [5] Only one copy of each gene is present [6] Within these H1 genes several single nucleotide polymorphisms (SNPs) have been reported (NCBI Single Nucleotide Polymorphism Database) To our knowledge, no acquired mutations have been reported to date in any human H1 gene Studies of the structure of H1 histones, their interaction with the nucleosome and their roles in controlling gene activity, indicate that these proteins have both an essential architectural function and an important task in regulating transcription [7,8] The precise functions of these multiple H1 subtypes and their modifications are not yet fully understood, but it has been reported that distinct H1 histone variants are preferentially localized to particular chromosomal domains [9–11] Although individual H1 subtypes are dispensable for mammalian development [12], it now seems clear that linker histones, in general, are essential for proper development [13] and it was recently found that one subtype, H1.2, had a specific role in DNA damage-induced apoptosis [14] It is useful therefore to examine the properties and expression of these variants as this furthers a better understanding of the relevance of this diversity for particular cellular activities The sequence similarity between histones H1.1–H1.5 requires highly efficient analytical methods for their resolution To date, the most widely utilized procedures for the study of human H1 proteins have been PAGE and low-pressure ion-exchange chromatography Two-dimensional gel electrophoresis allows the separation of several H1 variants [15,16], and four H1 subtypes were obtained by using BioRex 70 column chromatography [17–19] Both PAGE and lowpressure ion-exchange chromatography, however, are laborious, time-consuming and their resolution is unsatisfactory Recently, we described rapid and simple methods for the separation of rat and mouse H1 histones by using RP-HPLC [20,21] and high performance capillary electrophoresis (HPCE) [22–24] Furthermore, by applying hydrophilic interaction liquid chromatography (HILIC) excellent fractionations of various posttranslationally modified core [25,26] and linker [27,28] histones were obtained in both the analytical and the semipreparative scale 3674 B Sarg et al This article evaluates the potential utility of HILIC as a means of investigating the occurrence of sequence variations within linker histone subtypes from various human tumor cell lines By using HILIC we detected amino acid substitutions in H1.2 and H1.4 at the protein level In addition, sequencing of the corresponding gene segments confirmed these findings at the genome level Furthermore, we also screened DNA from 103 healthy individuals to obtain the allele frequency for these variants Results HILIC is an excellent technique for using to separate histone proteins and their modified forms [25–29] This study aimed to apply and optimize the HILIC technique in order to detect the occurrence of microsequence variations of H1 subtypes isolated from various human tumor cell lines As the level of histone H1 phosphorylation is lower in nondividing than in proliferating cells, we isolated H1 histones from cell cultures in the stationary phase, thus making it possible to reduce the occurrence of additional peaks caused by phosphorylated forms of the parent proteins A typical separation pattern using a PolyCAT A column and a two-step sodium perchlorate gradient (0–0.68 m) in the presence of 70% (v ⁄ v) acetonitrile and 0.015 m triethylammonium phosphate (TEAP) (pH 3.0) is shown in Fig 1A The CCRF-CEM H1 sample was separated into four peaks Analysis of several other cell lines showed the same pattern, but the relative concentrations of the subtypes varied (data not shown) H1 samples from Raji (Fig 1B) and K562 (Fig 1C) cells, however, showed different patterns, and additional peaks were detected, namely peak 3a in Fig 1B and peak 1a in Fig 1C To identify the individual peaks of the chromatograms we digested the various subfractions with chymotrypsin, a protease that specifically hydrolyzes peptide bonds at the C terminus of Tyr, Phe and Trp As human H1 histones contain only one Phe residue, it was expected that cleavage with chymotrypsin should produce two peptide fragments In fact, when the digested proteins were separated by RP-HPLC, two main fractions were obtained An example is shown in Fig 2, where a digest of fraction isolated by HILIC separation of H1 from CCRF-CEM cells (Fig 1A) was analyzed by RP-HPLC The purity and homogeneity of the two fractions was assessed by capillary electrophoresis (CE) (data not shown) To identify the fractions, amino acid sequencing was performed Fraction contained the C-terminal region starting after Phe at amino acid 105 No sequence data were FEBS Journal 272 (2005) 3673–3683 ª 2005 FEBS B Sarg et al H1 microsequence variants in human cell lines A B C Fig Hydrophilic interaction liquid chromatography (HILIC) separation of H1 histones isolated from human tumor cell lines H1 histone samples from (A) CCRF-CEM cells, (B) Raji cells, and (C) K562 cells were analyzed on a PolyCAT A column (4.6 mm · 250 mm) at 23 °C, and at a constant flow of 1.0 mLỈmin)1, by using a two-step gradient starting at solvent A ⁄ solvent B (100 : 0) [solvent A: 70% (v ⁄ v) acetonitrile, 0.015 M triethylammoniumphosphate (TEAP, pH 3.0); solvent B: 70% (v ⁄ v) acetonitrile, 0.015 M TEAP (pH 3.0) and 0.68 M NaClO4] The concentration of solvent B was increased from to 80% (v ⁄ v) during a time-period of and from 80 to 100% (v ⁄ v) during a time-period of 60 The isolated protein fractions (designated 1–4) were desalted by using RP-HPLC FEBS Journal 272 (2005) 3673–3683 ª 2005 FEBS Fig RP-HPLC analysis of peptide fractions of chymotrypsindigested H1 from CCRF-CEM cells Peak from H1 histones isolated with hydrophilic interaction liquid chromatography (HILIC) (Fig 1A) was digested with chymotrypsin, as described in the Experimental procedures The digest (containing  100 lg of protein) was injected onto a Nucleosil 300-5 C18 column (250 mm · mm) Analysis was performed at a constant flow of 0.35 mLỈmin)1 with a multistep acetonitrile gradient starting at solvent A ⁄ solvent B (85 : 15) (solvent A: water containing 0.1% (v ⁄ v) trifluoroacetic acid; solvent B: 85% (v ⁄ v) acetonitrile and 0.1% (v ⁄ v) trifluoroacetic acid) The concentration of solvent B was increased linearly from 15 to 23% during a time-period of 25 min, from 23 to 70% during a time-period of 45 and from 70 to 100% during a time-period of obtained from fraction because the first amino acid was blocked, indicating that fraction consisted of the N-terminal region of this H1 subtype By using protein sequence data, the four H1 peaks from CCRF-CEM cells (Fig 1A) were identified as follows: peak 1, histone H1.2 (SWISS-PROT P16403); peak 2, histones H1.3 (SWISS-PROT P16402) + H1.5 (SWISS-PROT P16401); peak 3, histone H1.4 (SWISS-PROT P10412); and peak 4, histone H1.5 Identification of the Raji H1 fractions (Fig 1B) yielded the same result for peaks 1, and 4, while peaks 3a and 3b contained histone H1.4 In order to exclude the presence of phosphorylated H1.4, which would be a reasonable explanation for this diversity, the Raji H1 proteins were incubated with alkaline phosphatase and subjected to HILIC Peak showed a dramatic decrease in size, whereas the two H1.4 peaks were not affected by the phosphatase (data not shown) Therefore, peak was identified as a phosphorylated form of H1.5 Surprisingly, the two H1.4 subfractions (Fig 1B, peaks 3a and 3b) were not separated either by HPCE or RP-HPLC or by different gel-electrophoretic methods (data not shown) To examine the structural difference between these two 3675 H1 microsequence variants in human cell lines proteins, we analyzed the two main fractions and obtained from chymotrypsin digestion and RP-HPLC separation (as shown for CCRF-CEM cells in Fig 2) of peaks 3a and 3b under HILIC conditions We found that fraction derived from peaks 3a and 3b had the same elution time, while fraction clearly differed Therefore, the C-terminal domain should be responsible for the microheterogeneity of histone H1.4 In order to elucidate the nature of this alteration, the C-terminal peptides were further cleaved with endoproteinase Glu-C Digests were analyzed by RP-HPLC by using a Nucleosil 300-5 C18 column (250 · mm; Fig 3) Fragmentation yielded three main peptides (I–III), which were identified by Edman degradation: fraction I (eluting at 27 min) consisted of a peptide starting at residue 105, fraction II (43 min) consisted of a peptide starting at residue 115, and fraction III (60 min) consisted of a peptide starting at residue 150 The purity of the fractions obtained was confirmed by CE (data not shown) Again, all peptides were analyzed under HILIC conditions, and the result showed that only the two fractions III contained peptides with different elution times Further peptide Fig RP-HPLC analysis of peptide fractions of endoproteinase Glu-C-digested fraction from Raji H1.4 Peak 3a from Raji H1 histones isolated with hydrophilic interaction liquid chromatography (HILIC) (Fig 1B) was digested with chymotrypsin and isolated by RP-HPLC (Fig 2) The fraction obtained was further digested with endoproteinase Glu-C, as described in the Experimental procedures, and the digest (containing  30 lg of protein) was injected onto a Nucleosil 300-5 C18 column (250 mm · mm) Analysis was performed at a constant flow of 0.35 mLỈmin)1 with a multistep acetonitrile gradient starting at solvent A ⁄ solvent B (95 : 5) [solvent A: water containing 0.1% (v ⁄ v) trifluoroacetic acid; solvent B: 85% (v ⁄ v) acetonitrile and 0.1% (v ⁄ v) trifluoroacetic acid] The concentration of solvent B was increased linearly from to 20% during a time-period of 65 and from 20 to 100% during a time-period of 25 3676 B Sarg et al sequencing of fraction III revealed that the two H1.4 HILIC peaks differed from one another by a single amino acid substitution – lysine at position 173 (peak 3b) was replaced by arginine (peak 3a) This result was further confirmed by subjecting the two H1.4 subfractions obtained by HILIC (Fig 1B) to electrospray ionization mass spectrometry analysis For peak 3a we found a mass of 21802.4 Da and for peak 3b a mass of 21774.8 Da, the latter being in close agreement with the wild-type H1.4 mass, which was calculated to be 21776.1 Da The mass difference observed between the two peaks was 27.6 Da, and this corresponds to the mass difference between lysine and arginine This microsequence variant H1.4 Lys173Arg, found in Raji cells in the same concentrations as the wild-type H1.4, was detected neither in CCRF-CEM and K562 cells nor in several other human cell lines (e.g U937, HL60) or in human tissue (placenta, testis) A further microheterogeneity was found when analyzing the K562 H1 sample by HILIC (Fig 1C) Peak 1, which was a single fraction in CCRF-CEM and Raji cells, and identified as histone H1.2, was separated into two subfractions (peaks 1a and 1b) in K562 cells RP-HPLC separation after chymotrypsin digestion of the two fractions 1a and 1b yielded two main fragments each (similar to Fig 2) HILIC analysis revealed that fraction from 1a had a shorter elution time than did fraction from 1b, whereas no differences were observed between the fraction samples Further cleavage of the fraction samples with endoproteinase Glu-C followed by peptide sequencing showed that the proteins differed by one amino acid out of a total of 212: peak 1a contained valine in position 17, while peak 1b contained wild-type alanine Histone H1.2 from all cell lines and tissues investigated contained only alanine at this location To confirm the H1.2 Ala17Val sequence variation, a 183 bp PCR product was amplified in the 5¢-UTR, and the start of the coding sequence of the H1.2 gene corresponding to the N-terminal tail of the protein PCR products from K562, Raji and wild-type blood donors were sequenced K562 DNA contained a homozygous CfiT substitution at nucleotide position 578 in the H1.2 gene, resulting in a change, in codon 18, from GCC to GTC (Fig 4), encoding alanine and valine, respectively This change in codon 18 corresponds to the substitution at amino acid position 17 in the H1.2 protein, as the initiating methionine is removed after translation in all H1 histones Traces of the wild-type C in position 578 were also detected (Fig 4) To obtain the population frequency of the H1.2 g578 CfiT substitution, 103 healthy individuals were screened by using a restriction fragment length FEBS Journal 272 (2005) 3673–3683 ª 2005 FEBS B Sarg et al H1 microsequence variants in human cell lines Fig DNA sequencing of H1.2 PCR products Fig DNA sequencing of H1.4 PCR products A B C Fig Restriction fragment length polymorphism (RFLP) analysis of H1.2 PCR products Lanes and 2, blood donors, heterozygous for g578 CfiT Lane 3, wild-type blood donor Lane 4, K562, homozygous for g578 CfiT Lane 5, uncleaved control Lane 6, 100 bp ladder polymorphism (RFLP) assay The wild-type H1.2 PCR product was cleaved into three fragments by BsuRI, while a PCR product containing the g578 CfiT substitution was cleaved into two fragments (Fig 5) We found 10 individuals to be heterozygous and two homozygous for the g578 CfiT substitution, resulting in an allele frequency of 6.8% in this population The presence of the g578 CfiT substitution in samples displaying the 130 bp fragment in the RFLP analysis was confirmed by DNA sequencing To detect the H1.4 Lys173Arg substitution, a 217 bp fragment of the H1.4 gene, corresponding to the C terminus of the H1.4 protein, was amplified by PCR Raji, K562 and wild-type blood donor PCR products were subjected to DNA sequencing Raji cells were heterozygous for an AAA to AGA codon change at position 174 (Fig 6), resulting in a Lys173Arg substitution in histone H1.4 To determine the allele frequency of this g1250 AfiG substitution in the H1.4 gene, a denaturating HPLC method was developed (Fig 7) Heteroduplex and mutant homoduplex formation was FEBS Journal 272 (2005) 3673–3683 ª 2005 FEBS Fig Denaturing HPLC of PCR-amplified H1.4 fragments (A) Wild-type blood donor (B) Raji, heterozygous for g1250 AfiG (C) Raji and wild-type PCR products in a : ratio resolved after mixing all PCR products with wild-type PCR products of H1.4 No G alleles were detected in the normal population of 206 alleles studied As the Raji cell line was derived from an individual suffering from Burkitt’s lymphoma, six other cell lines established from Burkitt’s lymphoma were screened for the g1250 AfiG substitution However, none of these cell lines contained this sequence variation Discussion A major question concerning the expression of individual H1 subtypes is whether they have coevolved with functional differences Although the precise function of H1 isoforms has yet to be determined, several observations suggest distinct and nonoverlapping roles for individual H1 variants Estimation of the rates of nucleotide substitution for mammalian H1 subtypes 3677 H1 microsequence variants in human cell lines H1a–H1e during evolution showed evidence of their functional differentiation [30] The study revealed that the rates of nucleotide substitution differed not only among subtypes, but also among domains For all subtypes, the synonymous substitution rate greatly exceeded the nonsynonymous rate, and the terminal domains were more variable than the central globular domain We detected two nonsynonymous substitutions in human H1 histones One caused alanine to be substituted by valine in the N-terminal region (position 17) of histone H1.2 in K562 cells, leading to two subfractions using HILIC The H1.2 Ala17Val variant constituted the major fraction of the H1.2 protein extracted, while a minor fraction of wild-type H1.2 was present In agreement with this, K562 DNA was found to carry a homozygous GCC to GCT change at codon 18 in the H1.2 gene There were remains of the wild-type C at g578 (Fig 4), explaining the minor wild-type H1.2 peak in the HILIC chromatogram This result is probably explained by the nondiploid karyotype of the K562 cells The H1.2 g578 CfiT sequence variant was also found to be present in a normal population, with the allele frequency 6.8%, and is therefore concluded to be a polymorphism The H1.2 Ala17Val was expressed in K562 cells and therefore probably also in normal individuals carrying this gene variant However, H1 histones show a high redundancy in knockout organisms, and the deletion of one or more subtypes causes increased expression of those remaining [12,31] Therefore, it cannot be completely ruled out that normal individuals carrying the H1.2 g578 CfiT are devoid of Ala17Val H1.2 expression This polymorphic site in the H1.2 protein has previously been recognized in human spleen [18] The corresponding SNP was recently reported (NCBI SNP database refSNP ID rs 2230653) In addition, three other SNPs have been reported in H1.2, all leading to synonymous changes (NCBI SNP) The role of the N-terminal tail of H1 histones is unclear, but is believed to be involved in positioning of the globular domain on the nucleosome [32] The N- and C-terminal tails of histone H1 adopt a random coil in solution [33] On binding of histone H1° to DNA, significant parts of the N terminus are likely to take on an a-helical structure [34], and this is probably also the case for other H1 subtypes As the tails of H1 histones not adopt their native conformation until they bind to chromatin, it is hard to predict the structural changes that a single amino acid substitution may trigger Substitution of valine for alanine may affect the predicted a-helical structure of the tail as valine has a less stabilizing effect on an a-helical struc3678 B Sarg et al ture If the structure is affected by the polymorphism, the positioning of H1.2 on the nucleosome, or the binding of H1.2 to chromatin, may be affected A further nonsynonymous substitution prompted the replacement of lysine with arginine in the C-terminal tail (position 173) of histone H1.4 in Raji cells By using HILIC, histone H1.4 was separated into two peaks: one wild-type H1.4; and one Lys173Arg H1.4 This microsequence variant was found for the first time and was present in stationary-phase cells in similar amounts as wild-type H1.4 Genetic analysis of Raji cells showed a heterozygous H1.4 g1250 AfiG substitution, prompting alteration of codon 174 from AAA to AGA, in agreement with the Lys173Arg substitution This sequence variant was not present in the 103 normal individuals that were screened, or in six other Burkitt’s lymphoma cell lines, implying that the Lys173Arg substitution is probably a mutation or a rare polymorphism detected, thus far, only in Raji cells Denaturing HPLC, used to screen for this genetic variant, provides a sensitive and highly specific method for investigating sequence variations [35], and the possibility of false negative results is unlikely Histone H1.4 mRNA (GenBank NM_005321) has been reported to contain a different polymorphism (NCBI SNP refSNP ID rs2298090), c455 AfiG, causing a Lys152Arg substitution (NP_005312.1) The C-terminal tail of histone H1 is believed to be responsible for the condensation of chromatin [32,36], and the condensing property of rat H1d probably resides in a C-terminal 34 amino acid stretch [37] Different H1 subtypes may have different chromatincondensing properties [38,39] On binding to DNA, the C-terminal tail probably adopts the structure of a segmented a helix [40] Replacement of lysine with arginine may affect the secondary structure of the C-terminal tail and the binding of H1.4 to chromatin, as arginine offers additional hydrogen-bonding abilities to DNA as compared to lysine Lysine 173 in H1.4 is situated in an SPKK motif, one of the known sites for H1 phosphorylation As far as we know, there are no differences in the phosphorylation behavior of SPKK and SPRK motifs Identification of numerous linker histone variants in vertebrates suggests that these proteins may play specialized roles Recent investigations using gene-targeting techniques, however, suggest that the specific timing of expression may have a greater functional significance than the nature of the individual H1 subtypes [2,41] It is, however, not clear to what extent the function of H1 variants depends on their primary sequence or on the specific timing of their expression The literature presents arguments in favor of both possibilities [2,7] FEBS Journal 272 (2005) 3673–3683 ª 2005 FEBS B Sarg et al The selective effect of linker histones on transcription of individual genes was demonstrated by using an in vivo system for inducible overexpression of different H1 subtypes (H1°, H1c) in mouse cells [42,43] Subtype-specific effects were shown to be related to differences in the structure of the globular domains [44] As H1 subtypes differ not only in primary sequence but also in turnover rate and extent of phosphorylation, they have the potential to add a great deal of flexibility to chromatin structure and transcriptional activation Both genes homologous to H1.2 and H1.4 have been disrupted in mouse, each resulting in viable and fertile knockout mice [12], indicating that these individual subtypes are dispensable and that compensatory effects reside between the subtypes, thus keeping H1 stoichiometry intact Despite the dispensability of wild-type H1 subtypes, however, microsequence variants may have biological significance Recent evidence reveals a highly specific function for H1.2 in DNA damage-induced apoptosis [14] As the somatic H1 subtypes H1.1–H1.5 show a high degree of sequence conservation, such specificity must rely on subtle differences in amino acid sequence Therefore, as histone H1 is implicated in chromatin organization, cell differentiation, gene regulation and apoptosis, these processes may be affected by minor sequence variations, including SNPs In conclusion, we have demonstrated the remarkably high resolving power of HILIC by using this technique to separate sequence variants within human linker histone subtypes We were thus able to detect an Ala17Val substitution in histone H1.2 in K562 cells, as well as a Raji-specific H1.4 Lys173Arg sequence variation at the protein level These observations were confirmed at the genetic level The significance of these variations is unclear, but it seems increasingly clear that minor sequence variations in linker histones may affect important cellular functions in vivo Experimental procedures Chemicals Sodium perchlorate (NaClO4), trifluoroacetic acid and triethylamine were purchased from Fluka (Buchs, Switzerland) All other chemicals were purchased from Merck (Darmstadt, Germany), unless indicated otherwise Cell lines and culture conditions CCRF-CEM acute lymphoblastic leukemia cells, Raji cells (originally derived from patients with Burkitt’s lymphoma) and K562 erythroleukemic cells were cultured in RPMI- FEBS Journal 272 (2005) 3673–3683 ª 2005 FEBS H1 microsequence variants in human cell lines 1640 medium (Biochrom, Berlin, Germany) supplemented with 10% (v ⁄ v) fetal bovine serum, penicillin (60 lgỈmL)1) and streptomycin (100 lgỈmL)1) in the presence of 5% (v ⁄ v) CO2 The cells were seeded at a density of · 104 cellsỈmL)1 and harvested after days to accumulate cells in stationary phase Preparation of H1 histones Human cells (6–7 · 109) were collected by centrifugation (800 g for 10 min) H1 histones were extracted with perchloric acid (5%, w ⁄ v) according to the procedure of Lindner et al [28] HPLC The equipment used for HPLC consisted of two 114M pumps, a 421A system controller and a Model 165 UV-visible-region detector (Beckman Instruments, Palo Alto, CA, USA) The effluent was monitored at 210 nm and the peaks were recorded by using Beckman System Gold software HILIC Whole human H1 samples were analyzed on a PolyCAT A column (4.6 mm · 250 mm; lm particle pore size; 30 nm pore size; ICT, Vienna, Austria) at 23 °C, and at a constant flow of 1.0 mLỈmin)1, by using a two-step gradient starting at solvent A ⁄ solvent B (100 : 0) [solvent A: 70% (v ⁄ v) acetonitrile, 0.015 m TEAP, pH 3.0; solvent B: 70% (v ⁄ v) acetonitrile, 0.015 m TEAP (pH 3.0) and 0.68 m NaClO4] The concentration of solvent B was increased from to 80% (v ⁄ v) during a time-period of and from 80 to 100% (v ⁄ v) during a time-period of 60 The isolated protein fractions were desalted by using RP-HPLC Histone fractions obtained in this manner were collected and, after adding 0.01 m 2-mercaptoethanol, freeze-dried and stored at )20 °C RP-HPLC The peptides obtained by limited chymotrypsin digestion of human H1 histones were separated by using a Nucleosil 300-5 C18 column (250 mm · mm internal diameter; lm particle pore size; end-capped; Macherey-Nagel, Duren, ă Germany) Samples of  100 lg were injected onto the column Chromatography was performed within 70 at a constant flow of 0.35 mLỈmin)1 with a multistep acetonitrile gradient starting at solvent A ⁄ solvent B (85 : 15) [solvent A: water containing 0.1% (v ⁄ v) trifluoroacetic acid; solvent B: 85% (v ⁄ v) acetonitrile and 0.1% (v ⁄ v) trifluoroacetic acid] The concentration of solvent B was increased linearly from 15 to 23% during a time-period of 25 min, from 23 to 70% during a time-period of 45 and from 70 to 100% during a time-period of 3679 H1 microsequence variants in human cell lines Human H1 peptide fractions obtained by digestion with endoproteinase Glu-C were separated by using the same column and solvents as described above The concentration of solvent B was increased linearly from to 20% during a time-period of 65 and from 20 to 100% during a time-period of 25 Fractions obtained in this manner were collected and, after adding 20 lL of 2-mercaptoethanol (0.2 m), were lyophilized and stored at )20 °C B Sarg et al tion, DNA samples collected randomly from 103 normal individuals in south-east Sweden were screened The individuals were selected from a population register and were 22–77 years of age (mean age, 52 years; SD, 17 years; 47% men and 53% women) The design was approved by Linkoă ping University Hospital Ethical Committee DNA was extracted from the blood samples by using the QIAampÒ DNA Blood Maxi kit (Qiagen) PCR amplification and RFLP analysis of H1.2 Chymotrypsin digestion Histone H1 subfractions ( 100 lg), obtained from human cell lines by HILIC fractionation, were digested with a-chymotrypsin (EC 3.4.21.1) (Sigma type I-S, ⁄ 150, w ⁄ w) in 100 lL of 100 mm sodium acetate buffer (pH 5.0) for 30 at room temperature The digest was subjected to RP-HPLC Endoproteinase Glu-C digestion Histone H1.4 fractions ( 50 lg) were digested with Staphylococcus aureus V8 Protease (Boehringer Mannheim, Mannheim, Germany; : 20, w ⁄ w) in 50 lL of 50 mm NH4HCO3 buffer (pH 7.8) for h at 37 °C Histone H1.2 fractions ( 50 lg) were digested in 50 lL of 25 mm phosphate buffer (pH 7.8) for h at room temperature The digests were subjected to RP-HPLC Peptide sequencing Peptide sequencing was performed on an Applied Biosystems Inc (ABI, Foster City, CA, USA) Model 492 Procise protein sequenator Typically, 5–100 pm of a peptide sample was run for 3–40 cycles, as required for an unambiguous identification Mass-spectrometric analysis Determination of the molecular masses of the two histone H1.4 subfractions obtained by the HILIC run was carried out by electrospray ionization mass spectrometry using an LCQ ion trap instrument (ThermoFinnigan, San Jose, CA, USA) Samples (15 lg) were dissolved in 50% (v ⁄ v) aqueous methanol, containing 0.1% (v ⁄ v) formic acid, and injected into the ion source DNA samples Genomic DNA from various cell lines was extracted by using the DneasyTM tissue kit (Qiagen, Hilden, Germany) and examined for sequence variations in codon 18 of H1.2 and in codon 174 of H1.4 To obtain the frequency of the two polymorphisms in H1.2 and H1.4 in a normal popula- 3680 A 183 bp fragment of the H1.2 gene (HIST1H1C, GenBank X57129) was amplified by using the PCR primers 5¢-CCCAGGCGCTGCTTC-3¢ (nucleotides 469fi483 of the H1.2 gene) and 5¢-CTCTGACACCGGGGGAC-3¢ (nucleotides 651fi635 of the H1.2 gene) The PCR was performed with 50 ng of DNA in a 20 lL reaction, containing mm MgCl2, 0.025 mL)1 Taq DNA polymerase, 20 mm Tris ⁄ HCl, pH 8.4, 50 mm KCl, lm of each primer (all Life Technologies, Gaithersburg, MD, USA) and 200 lm of each dATP, dCTP, dGTP and dTTP (Amersham Pharmacia Biotech, Piscataway, NJ, USA) After an initial denaturation at 94 °C for min, amplification was performed for 35 cycles with denaturation at 94 °C for min, annealing at 57 °C for and extension at 72 °C for min, in a thermal cycler (PTC-200; MJ Research, Watertown, MA, USA) The reaction was completed with an extension step at 72 °C for RFLP analysis was carried out by digesting the 183 bp PCR product with 10 U BsuRI (HaeIII) (MBI Fermentas, St Leon-Rot, Germany), at 37 °C overnight BsuRI recognizes the sequence 5¢-GGCC3¢, and the PCR product from wild-type DNA was digested into three fragments of 21, 53 and 109 bp Digestion of the PCR product from DNA containing the g578 CfiT substitution in the recognition sequence produced two fragments, one of 53 bp and one of 130 bp The digested PCR products were analyzed on Tris ⁄ borate ⁄ EDTA-agarose gels containing 1% (w ⁄ v) agarose (BioRad, Hercules, CA, USA), 3% (w ⁄ v) NuSieve GTG Agarose (FMC BioProducts, Rockland, ME, USA) and ethidium bromide (0.5 lgỈmL)1) The fragments were visualized under UV transillumination and photographed by using a Polaroid camera PCR amplification of H1.4 and polymorphism detection by using denaturing HPLC A 217 bp fragment of the H1.4 gene (HIST1H1E, GenBank M60748) was PCR amplified by using the primers 5¢-GA AGAGCGCCAAGAAGACC-3¢ (nucleotides 1173fi1191 of the H1.4 gene) and 5¢-CTACTTTTTCTTGGCTGCCG (nucleotides 1389fi1370 of the H1.4 gene), using the same conditions as described above For mutation analysis, a denaturating HPLC system (WAVEÒ Nucleic Acid Frag- FEBS Journal 272 (2005) 3673–3683 ª 2005 FEBS B Sarg et al ment Analysis System; Transgenomic, Crewe, UK) was used All samples were mixed with wild-type PCR product, which had previously been subjected to DNA sequence analysis, in a : ratio to ensure detection of g1250 AfiG homozygous mutants Before analysis, the samples were denaturated at 95 °C for and then gradually cooled, by °CỈmin)1, until 25 °C was reached, to allow heteroduplex formation The optimal melting temperature for the fragment was calculated by using the wave software, and a temperature of 63.3 °C was used for analysis The flow rate was 0.9 mLỈmin)1 and the total run time 7.2 Samples (20 lL) were injected onto the DNA Sep Column at 54% buffer A (0.1 m triethylammonium acetate, pH 7.0) and 46% buffer B [0.1 m triethylammonium acetate, pH 7.0, and 25% acetonitrile (v ⁄ v)] and heteroduplexes were separated by using a gradient starting at 49% buffer A and 51% buffer B, and gradually increasing to 40% buffer A and 60% buffer B DNA sequencing Direct cycle sequencing of H1.2 and H1.4 PCR products was performed with the corresponding forward PCR primer, using Thermo Sequenase radiolabeled terminator cycle sequencing kit (USB Corporation, Cleveland, OH, USA), and labeling with 33P dideoxy nucleotides (Amersham Pharmacia Biotech), according to the manufacturers’ recommendations Prior to sequencing, the PCR products were purified and concentrated by using GFX PCR DNA and the Gel band purification kit (Amersham Pharmacia Biotech) The labeled products from the sequencing reaction were separated on 6% (w ⁄ v) polyacrylamide gels, containing m urea, in a gel apparatus (OWL) at 70 W constant power After electrophoresis, the gel was dried and exposed to X-ray film Acknowledgements We thank A Devich, A Molbaek and S Gstrein for their excellent technical assistance This work, as part of the European Science Foundation EUROCORES Programme EuroDYNA, was supported by funds from the Austrian Science Foundation (project I23-B03) and the EC Sixth Framework Programme under Contract no ERAS-CT-2003-980409 and in part by the Swedish Cancer Society References Lennox RW (1984) Differences in evolutionary stability among mammalian H1 subtypes Implications for the roles of H1 subtypes in chromatin J Biol Chem 259, 669–672 FEBS Journal 272 (2005) 3673–3683 ª 2005 FEBS H1 microsequence variants in human cell lines Khochbin S & Wolffe AP (1994) Developmentally regulated expression of linker-histone variants in vertebrates Eur J Biochem 225, 501–510 Helliger W, Lindner H, Grubl-Knosp O & Puschendorf ă B (1992) Alteration in proportions of histone-H1 variants during the differentiation of murine erythroleukaemic cells Biochem J 288, 747–751 Albig W, Drabent B, Kunz J, Kalffsuske M, Grzeschik KH & Doenecke D (1993) All known human H1 histone genes except the H1° gene are clustered on chromosome Genomics 16, 649–654 Tanaka Y, Kato S, Tanaka M, Kuji N & Yoshimura Y (2003) Structure and expression of the human oocytespecific histone H1 gene elucidated by direct RT-nested PCR of a single oocyte Biochem Biophys Res Commun 304, 351–357 Albig W & Doenecke D (1997) The human histone gene cluster at the D6S105 locus Hum Genet 101, 284–294 Wolffe AP, Khochbin S & Dimitrov S (1997) What linker histones in chromatin? Bioessays 19, 249–255 Alami R, Fan Y, Pack S, Sonbuchner TM, Besse A, Lin Q, Greally JM, Skoultchi AI & Bouhassira EE (2003) Mammalian linker-histone subtypes differentially affect gene expression in vivo Proc Natl Acad Sci USA 100, 5920–5925 Mohr E, Trieschmann L & Grossbach U (1989) Histone H1 in two subspecies of Chironomus thummi with different genome sizes: homologous chromosome sites differ largely in their content of a specific H1 variant Proc Natl Acad Sci USA 86, 9308–9312 10 Schulze E, Trieschmann L, Schulze B, Schmidt ER, Pitzel S, Zechel K & Grossbach U (1993) Structural and functional differences between histone H1 sequence variants with differential intranuclear distribution Proc Natl Acad Sci USA 90, 2481–2485 11 Parseghian MH, Clark RF, Hauser LJ, Dvorkin N, Harris DA & Hamkalo BA (1993) Fractionation of human H1 subtypes and characterization of a subtypespecific antibody exhibiting non-uniform nuclear staining Chromosome Res 1, 127–139 12 Fan Y, Sirotkin A, Russell RG, Ayala J & Skoultchi AI (2001) Individual somatic H1 subtypes are dispensable for mouse development even in mice lacking the H1(0) replacement subtype Mol Cell Biol 21, 7933–7943 13 Fan Y, Nikitina T, Morin-Kensicki EM, Zhao J, Magnuson 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 14 Konishi A, Shimizu S, Hirota J, Takao T, Fan Y, Matsuoka 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 3681 H1 microsequence variants in human cell lines 15 D’Incalci M, Allavena P, Wu RS & Bonner WM (1986) H1 variant synthesis in proliferating and quiescent human cells Eur J Biochem 154, 273–279 16 Mannironi C, Rossi V, Biondi A, Ubezio P, Masera G, Barbui T & D’Incalci M (1987) Histone H1° is synthesized by human lymphocytic leukemia cells but not by normal lymphocytes Blood 70, 1203–1207 17 Ohe Y, Hayashi H & Iwai K (1986) Human spleen histone H1 Isolation and amino acid sequence of a main variant, H1b J Biochem (Tokyo) 100, 359–368 18 Ohe Y, Hayashi H & Iwai K (1989) Human spleen histone H1 Isolation and amino acid sequences of three minor variants, H1a, H1c, and H1d J Biochem (Tokyo) 106, 844–857 19 Hohmann P (1980) Species- and cell-specific expression of H1 histones in tissue culture cells Arch Biochem Biophys 205, 198–209 20 Lindner H, Helliger W & Puschendorf B (1988) Separation of Friend erythroleukaemic cell histones and high-mobility-group proteins by reversed-phase high performance liquid chromatography J Chromatogr 450, 309–316 21 Lindner H, Helliger W & Puschendorf B (1990) Separation of rat tissue histone H1 subtypes by reversed-phase HPLC Identification and assignment to a standard H1 nomenclature Biochem J 269, 359–363 22 Lindner H, Wurm M, Dirschlmayer A, Sarg B & Helliger W (1993) Application of high performance capillary electrophoresis to the analysis of H1 histones Electrophoresis 14, 480–485 23 Lindner H, Helliger W, Dirschlmayer A, Talasz H, Wurm M, Sarg B, Jaquemar M & Puschendorf B (1992) Separation of phosphorylated histone H1 variants by high performance capillary electrophoresis J Chromatogr 608, 211–216 24 Lindner H, Helliger W, Dirschlmayer A, Jaquemar M & Puschendorf B (1992) High performance capillary electrophoresis of core histones and their acetylated modified derivatives Biochem J 283, 467–471 25 Lindner H, Sarg B, Meraner C & Helliger W (1996) Separation of acetylated core histones by hydrophilic interaction liquid chromatography J Chromatogr A 743, 137–144 26 Sarg B, Koutzamani E, Helliger W, Rundquist I & Lindner HH (2002) Postsynthetic trimethylation of histone H4 at lysine 20 in mammalian tissues is associated with aging J Biol Chem 277, 39195–39201 27 Lindner H, Sarg B, Hoertnagl B & Helliger W (1998) The microheterogeneity of the mammalian H1° histone Evidence for an age-dependent deamidation J Biol Chem 273, 13324–13330 28 Lindner H, Sarg B & Helliger W (1997) Application of hydrophilic interaction liquid chromatography to the separation of phosphorylated H1 histones J Chromatogr A 782, 55–62 3682 B Sarg et al 29 Mizzen CA, Alpert AJ, Levesque L, Kruck TP & McLachlan DR (2000) Resolution of allelic and nonallelic variants of histone H1 by cation–exchange– hydrophilic interaction chromatography J Chromatogr B Biomed Sci App 744, 33–46 30 Ponte I, Vidal-Taboada JM & Suau P (1998) Evolution of the vertebrate H1 histone class: evidence for the functional differentiation of the subtypes Mol Biol Evol 15, 702–708 31 Sirotkin AM, Edelmann W, Cheng GH, Klein-Szanto A, Kucherlapati R & Skoultchi AI (1995) Mice develop normally without the H1° linker histone Proc Natl Acad Sci USA 92, 6434–6438 32 Allan J, Mitchell T, Harborne N, Bohm L & CraneRobinson C (1986) Roles of H1 domains in determining higher order chromatin structure and H1 location J Mol Biol 187, 591–601 33 Bradbury EM, Cary PD, Chapman GE, Crane-Robinson C, Danby SE, Rattle HW, Boublik M, Palau J & Aviles FJ (1975) Studies on the role and mode of operation of the very-lysine-rich histone H1 (F1) in eukaryote chromatin The conformation of histone H1 Eur J Biochem 52, 605–613 34 Vila R, Ponte I, Collado M, Arrondo JL, Jimenez MA, Rico M & Suau P (2001) DNA-induced alpha-helical structure in the NH2-terminal domain of histone H1 J Biol Chem 276, 46429–46435 35 Xiao W & Oefner PJ (2001) Denaturing high performance liquid chromatography: a review Hum Mutat 17, 439–474 36 Allan J, Hartman PG, Crane-Robinson C & Aviles FX (1980) The structure of histone H1 and its location in chromatin Nature 288, 675–679 37 Bharath MM, Ramesh S, Chandra NR & Rao MR (2002) Identification of a 34 amino acid stretch within the C-terminus of histone H1 as the DNA-condensing domain by site-directed mutagenesis Biochemistry (Mosc) 41, 7617–7627 38 Khadake JR & Rao MR (1997) Condensation of DNA and chromatin by an SPKK-containing octapeptide repeat motif present in the C-terminus of histone H1 Biochemistry (Mosc) 36, 1041–1051 39 De Lucia F, Faraone-Mennella MR, Derme M, Quesada P, Caiafa P & Farina B (1994) Histoneinduced condensation of rat testis chromatin: Testis-specific H1t versus somatic H1 variants Biochem Biophys Res Commun 198, 32–39 40 Clark DJ, Hill CS, Martin SR & Thomas JO (1988) Alpha-helix in the carboxy-terminal domains of histones H1 and H5 EMBO J 7, 69–75 41 Khochbin S (2001) Histone H1 diversity: bridging regulatory signals to linker histone function Gene 271, 1–12 42 Brown DT & Sittman DB (1993) Identification through overexpression and tagging of the variant type FEBS Journal 272 (2005) 3673–3683 ª 2005 FEBS B Sarg et al of the mouse H1e and H1c genes J Biol Chem 268, 713–718 43 Brown DT, Alexander BT & Sittman DB (1996) Differential effect of H1 variant overexpression on cell cycle progression and gene expression Nucleic Acids Res 24, 486–493 FEBS Journal 272 (2005) 3673–3683 ª 2005 FEBS H1 microsequence variants in human cell lines 44 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 3683 ... arginine may affect the secondary structure of the C-terminal tail and the binding of H1.4 to chromatin, as arginine offers additional hydrogen-bonding abilities to DNA as compared to lysine... cell lines was extracted by using the DneasyTM tissue kit (Qiagen, Hilden, Germany) and examined for sequence variations in codon 18 of H1.2 and in codon 174 of H1.4 To obtain the frequency of the... human tumor cell lines By using HILIC we detected amino acid substitutions in H1.2 and H1.4 at the protein level In addition, sequencing of the corresponding gene segments confirmed these findings