Article Information

• PDF (1919 kb)
• Full Text with Large Figures
• Cited by in Scopus (1)

PubMed

• Articles by Ron M. Finn
• Articles by Juan Ausió

Related Articles

• ACF, an ISWI-Containing and ATP-Utilizing Chromatin Assembly...

  ACF, an ISWI-Containing and ATP-Utilizing Chromatin Assembly and Remodeling Factor Cell, Volume 90, Issue 1, 11 July 1997, Pages 145-155 Takashi Ito, Michael Bulger, Michael J Pazin, Ryuji Kobayashi and James T Kadonaga Summary We describe the purification and characterization of ACF, an TP-utilizing hromatin assembly and remodeling actor. ACF is a multisubunit factor that contains ISWI protein and is distinct from NURF, another ISWI-containing factor. In chromatin assembly, purified ACF and a core histone chaperone (such as NAP-1 or CAF-1) are sufficient for the ATP-dependent formation of periodic nucleosome arrays. In chromatin remodeling, ACF is able to modulate the internucleosomal spacing of chromatin by an ATP-dependent mechanism. Moreover, ACF can mediate promoter-specific nucleosome reconfiguration by Gal4-VP16 in an ATP-dependent manner. These results suggest that ACF acts catalytically both in chromatin assembly and in the remodeling of nucleosomes that occurs during transcriptional activation. Summary | Full Text | PDF (378 kb)

• Nucleosomes Unfold Completely at a Transcriptionally Active ...

  Nucleosomes Unfold Completely at a Transcriptionally Active Promoter Molecular Cell, Volume 11, Issue 6, 1 June 2003, Pages 1587-1598 Hinrich Boeger, Joachim Griesenbeck, J.Seth Strattan and Roger D. Kornberg Summary It has long been known that promoter DNA is converted to a nuclease-sensitive state upon transcriptional activation. Recent findings have raised the possibility that this conversion reflects only a partial unfolding or other perturbation of nucleosomal structure, rather than the loss of nucleosomes. We report topological, sedimentation, nuclease digestion, and ChIP analyses, which demonstrate the complete unfolding of nucleosomes at the transcriptionally active promoter of the yeast . Although nucleosome loss occurs at all promoter sites, it is not complete at any of them, suggesting the existence of an equilibrium between the removal of nucleosomes and their reformation. Summary | Full Text | PDF (490 kb)

• Physical characteristics of mouse sperm nuclei

  Physical characteristics of mouse sperm nuclei Biophysical Journal, Volume 16, Issue 7, 1 July 1976, Pages 811-825 A.J. Wyrobek, M.L. Meistrich, R. Furrer and W.R. Bruce Abstract The nuclei of epididymal sperm, isolated from C57BL/6J and CBA/J inbred mice by their resistance to trypsin digestion, retain the shape differences of the intact sperm head. Various physical characteristics of these nuclei were measured and compared. The measurement of the projected dimensions of nuclei showed that the CBA nuclei are 13.5% longer than C57BL/6 nuclei (8.64 +/- 0.02 mum compared with 7.61 +/- 0.02 mum), 0.8% narrower (3.51 +/- 0.01 vs. 3.54 +/-0.01 mum) with 6.8% more area (22.34 +/- 0.10 vs. 20.91 +/- 0.09 mum2). However, the volumes of the nuclei as based on reconstructing calibrated electronmicrographs of serial sections of the nuclei indicated that CBA are about 7% smaller than C57BL/6 nuclei (3.72 +/- 0.08 vs. 4.01 +/- 0.03 mum3). The buoyant density of the CBA nuclei is 1.435 +/- 0.002 g/cm3 compared with 1.433 +/- 0.002 g/cm3 for the C57BL/6 nuclei as determined on linear CsCl and Renografin-76 density gradients and confirmed by a technique utilizing physiological tonicities. Therefore, the average mass of the CBA nuclei is less than that of the C57BL/6 nuclei (5.34 +/- 0.12 vs. 5.75 +/- 0.05 pg). The sedimentation velocities at unit gravity of nuclei from 11 inbred strains differ over a range of more than 6% with CBA nuclei sedimenting about 2.0% more slowly than C57BL/6 nuclei. We show that for these nuclei the sedimentation velocity can be related to their buoyant density, volume and a sedimentation shape factor. Within the errors of our measurements of these various characteristics, it was found that C57BL/6 and CBA nuclei have similar sedimentation shape factors. Therefore, the difference in sedimentation velocity between these nuclei appears to be primarily a result of differences in volume. The possible applications of these techniques to the physical separation of sperm are evaluated in the discussion. Abstract | PDF (1103 kb)

• …more

Copyright © 2008 The Biophysical Society. All rights reserved.
Biophysical Journal, Volume 95, Issue 3, 1314-1325, 1 August 2008

doi:10.1529/biophysj.108.130021

Proteins

sNASP, a Histone H1-Specific Eukaryotic Chaperone Dimer that Facilitates Chromatin Assembly

Ron M. Finn^*, 1, Kristen Browne^*, 1, 2, Kim C. Hodgson^*, † and Juan Ausió^*, ,  

^* Department of Biochemistry and Microbiology, University of Victoria, Petch Building, Victoria, BC V8W 3P6, Canada
^† Department of Biology, University of Victoria, Petch Building, Victoria, BC V8W 3P6, Canada

Address reprint requests to Juan Ausió, Department of Biochemistry and Microbiology, University of Victoria, Victoria, BC V8W 3P6, Canada. Tel.: 250-721-8863; Fax: 250-721-8855.

¹ Ron M. Finn and Kristen Brown contributed equally to this work.

² Kristen Brown's present address is Zoology Department, University of British Columbia, 2350 Health Sciences Mall, Vancouver, BC V6T 1Z3, Canada.

Histones need to be assembled onto the newly synthesized DNA template during the S phase of the cell cycle 1, and they need to be disassembled and reassembled at other stages to facilitate and permit DNA accessibility during transcription 2, DNA recombination, and repair 3. This dynamic histone exchange is mediated by histone chaperones such as NAP-1** 4, which preferentially binds to H2A-H2B dimers 5, CAF-1/antisilencing factor 1, which binds to H3-H4 tetramers 6,7,8, and nucleoplasmin 9/nucleophosmin, which binds to both 10 (see Loyola and Almouzni 11 for a recent comprehensive review on core histone chaperones).

In contrast to the large amount of information on core histone chaperones, very few examples are known of linker histone chaperones, which remain largely uncharacterized. One such protein is NASP, a protein that shares a large degree of homology with the Xenopus histone-binding protein N1/N2 12,13. NASP binds linker histones that are not bound to DNA in vivo and, hence, may provide a potentially useful reagent for in vitro reconstitution of linker histones onto nucleosome arrays. NASP binds to linker histones, transporting them into the nucleus and exchanging them onto DNA 14,15. This chaperone is present in all mitotic cells, either as an sNASP or a tNASP 16,17. Alternative splicing of the pre-mRNA results in two mature mRNA forms corresponding to the full-length tNASP and the shorter sNASP 16. In humans, sNASP is a 449-amino-acid protein that is essential for cell proliferation, cell cycle regulation, and DNA replication 18,19. During normal cell cycle progression, the level of NASP expressed is under strict control 15,16. Overexpression of NASP leads to a delay in the progression through the G1/S border 15,16. NASP exists in a multichaperone nucleosome-remodeling complex that is composed of not only sNASP but CAF-1 (p150, p60, p48), H3.1, H4, Asfa, Asfb, checkpoint homolog 2, histone acetyltransferase 1, and importin 4 20,21. This complex coordinates the deposition of H3.1–H4 histones into nucleosomes and allows for the progression of DNA replication through phosphorylation signals at specific checkpoints 22. The implication of this is that completion of S phase in a normal cell cycle requires the presence of NASP 21,23,24. Also, NASP has been found to interact in vivo with Ku70/Ku80 and DNA protein kinase implying that NASP may play a role in DNA repair 14. Intriguingly, it has yet to be shown whether NASP binds DNA directly 22, and NASP sequence possesses motifs indicating possible interaction sites for several essential proteins such as checkpoint homolog 2 and p300 22.

The distribution of NASP over a broad spectrum of tissues (sNASP) and in testes (tNASP) has been well documented in mammals 16. Here we report that these two species of NASP exhibit a conserved wide distribution in eukaryotes and provide the first biophysical characterization of sNASP, which is shown to be a genuine linker histone chaperone.

All of the chemicals and reagents used were of molecular biology grade. HeLa cell RNA was extracted using Qiashredder and RNeasy Mini Kit (Qiagen, Mississauga, ON). cDNA synthesis was conducted using Invitrogen's First Strand cDNA Kit (Invitrogen, Carlsbad, CA). Restriction enzymes were purchased from New England Biolabs (NEB, Pickering, ON). Purification of DNA was done using Qiagen miniprep PCR purification and gel extraction kits. Sequencing was conducted at the DNA Sequencing Facility of Centre for Biomedical Research at the University of Victoria. Rabbit anti-NASP antibody was a gift from Dr. M. O’Rand (Dept. of Cell and Developmental Biology, University of North Carolina at Chapel Hill, Chapel Hill, NC) 16. Goat antirabbit antibodies were purchased from Abcam (Cambridge, MA). Micrococcal nuclease was obtained from Worthington Biochemical (Lakewood, NJ).

SDS-PAGE was performed according to Laemmli 25. Native (6.0%)-PAGE (acrylamide:bis-acrylamide 29:1 (w:w)) for nucleosome analysis was carried out in 20mM sodium acetate, 1mM EDTA, 20mM Tris-HCl (pH 7.2), E buffer as described elsewhere 26,27.

The entire coding sequence of human sNASP (GenBank accession number NM_152298 was amplified from HeLa cell cDNA using the following primers and cycle conditions: forward 5′-GCGAATTCGGATCCATGGCCATGGAGTCCACA-3′, reverse 5′-GCCTCGAGGAATTCTTAACATGCAGTGCTTTCAACTG-3′; 3min at 94°C then a touchdown PCR from 64°C to 52°C annealing temperature for 1 min decreasing 3°C every three cycles then holding at 52°C annealing temperature for 18 cycles. Melting temperatures of 94°C for 1 min and extension temperatures of 72°C for 2min were used, with a final extension of 10min at 72°C. PCR products were then ligated into pCR2.1 TOPO vector (Invitrogen) sequenced, then digested with BamHI and XhoI for ligation into pET28a His-tag vector (Novagen, EMD Chemicals, Darmstadt, Germany) for an N-terminal thrombin-cleavable His-tag.

pET28a-sNASP was grown in Escherichia coli Rosetta (DE3) competent cells (Novagen, EMD Chemicals) in TB broth plus 34μg/ml kanamycin at 37°C to an OD₆₀₀ of ∼0.6 then induced for 2h with 0.4mM isopropyl β-D-1-thiogalactopyranoside. The expressed product had a calculated molecular mass of 52.4kDa with the His-tag. For purification, cells were suspended in lysis buffer (20mM Tris-HCl (pH 8.0), 500mM KCl, 20mM imidazole, 5mM β-mercaptoethanol, 15% glycerol (w/v), 1mM EDTA (pH 8.0), 0.1% NP-40 (w/v), and 1/25 dilution of Complete Protease Inhibitor Tablet (Roche, Basel, Switzerland) (dissolved in 1mL dH₂O), sonicated, and centrifuged. The supernatant was adjusted to 2mM MgCl₂ and filtered through a 0.45-μm membrane. Ni Sepharose High-Performance beads (Invitrogen) were equilibrated with lysis buffer in a column, and the sample was then loaded at ∼5ml/min. The column was washed with 20ml of wash buffer (40mM Tris-HCl (pH 8.0), 0.5M NaCl, 1/50 Complete protease inhibitors dissolved in 1ml dH₂O (Roche)), 10ml of wash buffer plus 40mM imidazole, then the protein was eluted with 20ml of wash buffer plus 0.5M imidazole, and 5-ml fractions were collected. sNASP-containing fractions were confirmed by 10% SDS-PAGE, then dialyzed extensively into Buffer A (40mM Tris-HCl (pH 8.0)). Dialyzed fractions were loaded onto Q-HiTrap 5-ml anion-exchange columns (Amersham, Piscataway, NJ) using the AKTA-fast protein liquid chromatography system (Amersham). A gradient of 20–60% Buffer B (Buffer A plus 1M NaCl) was used to elute sNASP, which usually came out at ∼40%=400mM NaCl. sNASP-containing fractions were confirmed on 10% SDS-PAGE and then dialyzed extensively against 20mM Tris-HCl (pH 8.0). In several instances, when indicated, the His-tag was removed by thrombin cleavage using a Novagen thrombin kit (Novagen) following the manufacturers’ instructions.

Native histone H1, H2A-H2B dimers, and H3-H4 tetramers were purified from HeLa cells, and chicken erythrocyte histone H1 and H5 were obtained from chicken blood. Chromatin obtained from these tissues 28,29 was loaded onto a hydroxyapatite column, and histones were eluted with a NaCl gradient in 10 mM phosphate buffer (pH 6.7) as described elsewhere 30. The trypsin-resistant core (comprising the winged helix domain) of HeLa H1 histones was obtained by digestion with trypsin in the presence of 2M NaCl plus 20mM Tris-HCl (pH 7.5) and purified by CMC-25 Sephadex as described by Jutglar et al. 31.

Tissue samples were pulverized in liquid nitrogen with a mortar and pestle. The resulting powder was suspended in lysis buffer (20mM Tris-HCl (pH 7.5) plus 1% SDS) and boiled for 1 min. Samples were then centrifuged for 5 min at 16,000×g. The pellet was discarded, and the supernatant was used directly for electrophoresis.

Tissue lysates were mixed with 2× loading buffer (125mM Tris-HCl (pH 6.8), 4% (w/v) SDS, 20% (w/v) glycerol, 1.72M β-mercaptoethanol, bromophenol blue) and boiled for 5min. Samples were then run on 10% SDS-PAGE at 100V. After electrophoresis, gels were transferred to polyvinylidene fluoride membranes at 100V for 2h in transfer buffer (192mM glycine, 25mM Tris-HCl, 20% (v/v) methanol). Blots were blocked for 30min at room temperature or overnight at 4°C in blocking buffer (5% (w/v) skim milk in phosphate-buffered saline plus 0.1% (v/v) Tween-20). Rabbit antibodies raised against recombinant mouse NASP 16 were diluted 1:1000 in blocking buffer. Secondary goat antirabbit antibodies (Abcam) were diluted 1:5000 in blocking buffer. Both primary and secondary incubations were done for 1h at room temperature with three 5-min washes with phosphate-buffered saline plus 0.1% Tween-20 in between and after incubations. Secondary antibodies were detected with enhanced chemiluminescence (GE Healthcare, Baie d’Urfe, QC) and exposure to x-ray film.

Recombinant sNASP in 20mM Tris-HCl (pH 7.5) buffer was digested with trypsin (EC 3.4.21.4) (type III) (Sigma-Aldrich, St. Louis, MO) for different amounts of time at room temperature using an E:S ratio of 1:1000 (mol:mol). The digests were analyzed by SDS-PAGE. Alternatively they were fractionated by HPLC using a C₁₈ Vydac column (GRACE-Vydac, Hesperia, CA) 32, and the fractions were analyzed by MALDI-time of flight (UVic-Genome BC Proteomics Centre, University of Victoria, Victoria, BC). In some instances the trypsin digests were blotted from the SDS-PAGE onto polyvinylidene fluoride membranes, and the membranes were stained with brilliant blue coomassie stain for a few minutes and air dried. The bands of interest were excised and subjected to N-terminal Edman degradation sequencing (Advanced Protein Technology Centre, Hospital for Sick Children, Toronto, ON).

CD analysis was carried out on a JASCO J720 (Jasco, Easton, MD) as described previously 33. The spectra were analyzed using a Contin LL 34 algorithm.

An extinction coefficient of 20,190M⁻¹ cm⁻¹ at 276nm was used for NASP as estimated from its amino acid sequence 35. The extinction coefficient of histone H5 was 48,000cm⁻¹ M⁻¹ at 230nm 36. For DNA, an extinction coefficient of 20cm² mg⁻¹ at 260nm was used 37.

These experiments were performed on an XL-I (Beckman-Coulter, Fullerton, CA) analytical ultracentrifuge. Sedimentation velocity experiments were carried out at 20°C and 44,000rpm in an An-55 aluminum rotor using double-sector aluminum-filled Epon centerpieces. Sedimentation equilibrium experiments were carried out at 4°C in an An-60 titanium rotor using double-sector aluminum-filled Epon centerpieces. The scans were analyzed by the method of van Holde and Weischet 38 using UltraScan version 9.3 39 sedimentation data analysis software (Borries Demeler, Missoula, MT). The partial specific volume of NASP (0.720cm³/g), and the frictional ratio f/f₀ were determined using the same software. Sedimentation velocity analysis was also used to determine the molecular mass using a Monte Carlo analysis 40,41.

Secondary structure prediction was performed using the hierarchical neural network protein sequence analysis 42. Tertiary structure prediction was carried out using the SWISS-MODEL server 43 and the Phyre server 44.

Chicken erythrocyte chromatin was prepared as described elsewhere 28. Long chromatin fragments were fractionated using a 5–20% sucrose gradient 45. A heterogeneous fraction with a 23 nucleosome average number was thus obtained. Approximately half of this fraction was stripped of linker histones using CM C-25 Sephadex (GE Healthcare) as described previously 46. The linker histone-containing and -depleted counterparts were analyzed by sedimentation velocity dialysis against different NaCl concentrations in 10mM Tris-HCl (pH 7.5), 0.2mM EDTA buffer.

Linker histone-depleted nucleosome arrays in 50mM NaCl, 10mM Tris-HCl (pH 7.5), 0.1mM EDTA were directly mixed with chicken erythrocyte histone H5 in the same buffer to a ratio of ∼2 mol H5:1 mol nucleosome. The mixture was incubated for 30min at room temperature, and aliquots were next dialyzed against different NaCl concentrations in 10mM Tris-HCl (pH 7.5), 0.2mM EDTA buffers and analyzed by sedimentation velocity.

Histone H5 and NASP were mixed at a ratio of ∼1:1.2mol:mol in 50mM NaCl, 10mM Tris-HCl (pH 7.5) buffer and incubated for 5min at room temperature. The H5-NASP mixture was then added to the linker histone-depleted nucleosome arrays at a ratio of 2mol H5:1 mol nucleosome and treated and analyzed as for direct mixing. In some instances (see the results), after incubation, the samples were brought to 4mM MgCl₂ and centrifuged at maximum speed in an Eppendorf (Hamburg, Germany) microcentrifuge at 4°C, and the pellet containing the H1-reconstituted chromatin was resuspended in 10mM Tris-HCl, 0.2mM EDTA and used for further analysis 47.

Histone H1-reconstituted complexes at different NaCl concentrations in 10mM Tris-HCl (pH 7.5) buffer with 0.05mM CaCl₂ were digested with micrococcal nuclease at 37°C for different amounts of time. Thirty units of micrococcal nuclease per milligram of DNA was used in these digestions. After digestion, the samples were mixed with an equal volume of a solution containing 30% sucrose, 0.8% SDS, and a mixture of bromphenol blue and cyanol green, incubated at 37°C and loaded directly onto a 4.5% native-PAGE to visualize the DNA composition of the products of digestion.

NASP is widely distributed in mammals 48, where it has been shown to exist in a smaller sNASP isoform, ∼400–500 amino acids, and a larger tNASP isoform, which is ∼300 amino acids longer. Although the critical cellular functions assigned to this chaperone and the identification of its activity in Caenorhabditis elegans48 suggest that the protein should be evenly distributed throughout eukaryotes, a comprehensive analysis of its distribution across different taxonomic groups has been lacking. The analysis shown in Fig. 1 indicates that sequences with a high extent of sequence similarity to mammalian t/sNASP can be identified in both vertebrate and invertebrate organisms including fungi (FIGURE 1A). Furthermore, proteins with similar molecular mass to both tNASP and sNASP can be identified in chordates and more specifically in species representative of each one of the main vertebrate classes from fish to mammals (FIGURE 1B). In all instances the apparent molecular masses in SDS-PAGE of the somatic and testicular forms are larger than the canonical values from the sequences, a fact that reflects the highly negatively charged nature of these proteins.

Display large version of this figure

FIGURE 1

NASP is widely distributed throughout eukaryotes. (A) Alignment of NASP amino acid sequences of several representative invertebrate and vertebrate organisms. The sequences and Genebank accession numbers are: 1), NM_172164 Homo sapiens (human) isoform 1; 2), NM_152298 Homo sapiens (human) isoform 3; 3), AF349432 Mus musculus (mouse); 4), XM-001235059〈removed hyperlink field〉 Gallus gallus (rooster); 5), NM_001088068 histone-binding protein N1/N2 Xenopus laevis (African clawed frog); 6), CAA28419 Xenopus laevis (African clawed frog); 7), BC068344 Danio rerio (zebrafish); 8), Ensembl Gene Report ENSCINT00000019618 Ciona intestinalis (sea squirt tunicate); 9), XP_791433 Strongylocentrotus purpuratus (sea urchin); 10), XP_970862 Trilobium castaneum (red flour beetle); 11), NM_063979 Caenorhabditis elegans (nematode) isoform 1; 12), XP_961229 Neurospora crassa (red bread mold). (B) Zoowestern analysis of NASP. (lane 1) Ciona intestinalis (mantle); (lane 2) Squalus acanthias (spiny dogfish, heart); (lane 3) S. acanthias (spiny dogfish, testes); (lane 4) X. laevis (frog, liver); (lane 5) Alligator mississippiensis (alligator, liver); (lane 6) G. gallus (chicken, liver); (lane 7) G. gallus (chicken, testes); (lane 8) M. musculus (mouse, liver); (lane 9) M. musculus (mouse, testes).

Importantly, in vertebrates, all the sequences identified as putative NASP bear homology to Xenopus laevis N1/N2 histone-binding protein (FIGURE 1A, line 5) 16,49. N1/N2 and nucleoplasmin are two histone chaperones 50 that are present in the oocytes and eggs of Xenopus, where they are found preferentially associated with core histones H3-H4 and H2A-H2B, respectively 51,52 (see recent reviews 10,50). In comparison to nucleoplasmin, N1/N2 has been less extensively characterized, and the molecular details of its histone chaperone activity and its involvement in chromatin assembly/remodeling still require elucidation.

To characterize the structure of human sNASP, we expressed and purified the recombinant version of this protein. FIGURE 2A-B shows an electrophoretic analysis of the protein thus obtained. In contrast to nucleoplasmin, which forms pentameric complexes that are highly stable in solution in the presence of SDS when not heated 53, recombinant human sNASP remains as a monomer in the presence of SDS with an apparent molecular mass of 60–70kDa. As in the case of the native form (FIGURE 1B), the apparent molecular mass of the recombinant sNASP is higher than the 52.4kDa estimated from the sequence of the recombinant version.

Display large version of this figure

FIGURE 2

Electrophoretic characterization of the recombinant human sNSAP used in this article. (A) SDS (15%)-PAGE of increasing loading amounts (triangles) of purified recombinant sNASP. M is (NEB, broad-range prestained) protein marker. (B) SDS (10%)-PAGE of increasing amounts of human recombinant sNASP before (−) and after (+) treatment with thrombin to remove the N-terminal His-tag. (C) SDS (10%)-PAGE of increasing amounts (triangles) of recombinant sNASP in comparison to recombinant X. laevis nucleoplasmin (N) 85. The proteins were dissolved in SDS sample buffer and loaded directly onto the gel without heat treatment (unboiled, N-U, NASP-U) or previously heated at 100°C for 5min before loading onto the gel (boiled, N-B, NASP-B). M* is (Bio-Rad Precision Plus Protein Standard) protein marker.

Next we determined the secondary structure of the protein using CD. FIGURE 3A shows the CD spectrum obtained in this way. Fitting of the spectrum using the Contin algorithm 34 revealed that the protein consists of 54% helix and 46% random coil. These values are in very good agreement with the secondary structure of 51% helix, 43% random coil, and 6% extended strand estimated using the Hierarchical Neural Network prediction analysis 42 and with the observations reported earlier using a similar kind of predictive analysis 49.

Display large version of this figure

FIGURE 3

Determination of the secondary structure of sNASP. (A) CD spectrum. The spectrum was recorded at 20°C in 100mM NaCl, 40mM Tris-HCl (pH 8.0) buffer. (B) Secondary structure prediction using the hierarchical neural network protein sequence analysis 42.

The tertiary and quaternary structures of the protein were analyzed in the analytical ultracentrifuge using both sedimentation velocity and sedimentation equilibrium experiments. The results of such analyses are summarized in Fig. 4. FIGURE 4A shows the integral distribution of the sedimentation coefficient of human sNASP under different ionic strengths, in the range from 25 to 500mM NaCl, showing that the sedimentation behavior is little affected by the salt concentration in this range. FIGURE 4B shows the result of the Monte Carlo analysis 41 of the sedimentation velocity in 25mM NaCl. This type of analysis not only allows an accurate analysis of the sedimentation coefficient but also provides a solution to the Lamm equation and allows the determination of the molecular masses of the different macromolecules present in the sample. It was determined that the major species present had an average sedimentation coefficient of s_20,w=4.5±0.2S corresponding to a molecular mass of 100,000 Daltons. The same sedimentation coefficient was obtained when the N-terminal His-tag of the protein was removed (FIGURE 4A, inset). A similar value of 99,000 Daltons was obtained from the best fits to the sedimentation equilibrium profiles of the sample under the same ionic conditions (FIGURE 4C). Thus, under the different ionic strengths used here, the protein remained as a dimer.

Display large version of this figure

FIGURE 4

Analytical ultracentrifuge analysis of recombinant human sNASP. (A) Integral distribution of the sedimentation coefficient of sNASP at different ionic strengths: 25mM NaCl (black triangles); 150mM NaCl (black circles); 500mM NaCl, 20mM Tris-HCl (pH 7.5) buffer (black squares). The inset shows the integral distribution of the sedimentation coefficient of a sNASP sample, before (white squares) and after (black circles) treatment with thrombin to remove the N-terminal His-tag, analyzed in 150mM NaCl, 20mM Tris-HCl (pH 7.5) buffer. The sedimentation velocity runs were performed at 44,000rpm and at 20°C. (B) Monte Carlo analysis of the sedimentation velocity experiment shown in A at 25mM NaCl. (C) Sedimentation equilibrium analyses of sNASP at different salt concentrations. The top plot shows the absorbance at 230nm as a function of the square of the radial distance of the sample at any position within the cell (r) minus the square of the radial position at a reference (r(ref)). The continuous line in this plot was obtained by fitting the experimental data (circles) to a single ideal species with a Mr of 99,000 Da. The bottom plot shows χ² residuals as a function of (r² − r(ref)²) for the best fit (solid line).The data shown were obtained at 17,000rpm and at 4°C.

When the value of the sedimentation coefficient s_20,w=4.5 was used in conjunction with the molecular mass of the dimer, using as an average preferential hydration parameter of the protein the value of 0.22g water/g protein 54, a frictional coefficient ratio f/f₀=1.68 was determined as described elsewhere 55. This corresponds to a molecule with a prolate ellipsoid shape of axial ratio a/b=12. The highly asymmetric value thus obtained suggests that the two monomers of sNASP are held together in a head-to-tail type arrangement with the dimer adopting a highly extended conformation.

Further insight into the folding of the human sNASP molecule was obtained by trypsin digestion. The results of such analysis are depicted in FIGURE 5A–C. The pattern of digestion obtained (FIGURE 5A) with this protease was very similar regardless of whether the digestion was carried out in the presence of 10mM NaCl (FIGURE 5AB, lane 3) or at 2M NaCl (FIGURE 5B, lane 2 and results not shown). The main tryptic-resistant peptides were identified by a combination of HPLC/SDS-PAGE/MALDI and N-terminal sequencing approach (see Materials and Methods). To help visualize the location of the trypsin-resistant domains determined in this way, a three-dimensional prediction analysis of the NASP monomer was carried out using the SWISS-MODEL server 43 (FIGURE 5C, ID). As can be seen in FIGURE 5CD, the main trypsin-resistant peptides (1, 2, and 3 in FIGURE 5CIII) are defined by the three predicted tetratricopeptide repeat domains and the coiled-coil domain. The previously defined histone-binding domains are contained within trypsin-resistant peptide 1. Interestingly, binding of histone H1 results in the preferential protection of peptide 5, helping us identify a novel histone-binding site that encompasses the region between amino acids 280 and 300 of human sNASP (FIGURE 5CIV).

Display large version of this figure

FIGURE 5

Tertiary structure of sNASP. (A) SDS-PAGE analysis of the time course of digestion of NASP by trypsin (E:S 1:1000) in 100mM NaCl, 20mM Tris-HCl (pH 7.5). (lane 1) Undigested sample; (lanes 2–5) peptides generated after 5, 10, 20, and 40min of digestion, respectively. (B) SDS-PAGE analysis of NASP (lane 1) and a NASP-HeLa histone H1 (1 mol:1 mol) complex (lane 4) before (lanes 1 and 4) and after digestion with trypsin (lanes 2, 3, and 5). Digestion was for 20min under the same conditions as in A (lanes 3 and 5) or in 2M NaCl, 20mM Tris-HCl (pH 7.5) (lane 2). CE, chicken erythrocyte histone marker; M, protein marker. The boxes highlight the peptides subjected to N-terminal sequencing analysis. (C) (I) Secondary structure organization derived from the tertiary structure prediction of sNASP using the Phyre program 44 in comparison to (II) the α-helix structure predicted from the hierarchical neural network analysis (see FIGURE 3B). Also shown are (III) the distribution of the trypsin-resistant peptides identified by N-terminal Edman degradation sequencing and MALDI-time of flight as well as (IV) the sites of histone H1 binding (orange boxes, 115–128, 138–172) described by Batova and O’Rand 70 or identified here (orange box, 280–300) 4. The numbers in brackets in III correspond to the peptide numbers identified in B. TPR, tetratricopeptide repeat domain; CCD, coiled-coil domain. (D) Tertiary structure predicted from the primary structure of sNASP using the Phyre program 44.

Fig. 6 shows the characterization of the interaction of human sNASP with core and linker histones using native gel electrophoresis. FIGURE 6A shows an SDS-PAGE electrophoretic analysis of the samples used for this analysis. As seen in FIGURE 6B, NASP forms discrete complexes at a ratio of 2mol of H1/mol NASP dimer. In contrast, H2A-H2B dimers or H3-H4 tetramers bind to NASP forming large aggregates that do not enter the native gels. Thus, it seems that NASP exhibits binding specificity to linker histones, and although it is able to bind to core histones, it does so in a less specific way. The nonspecific binding is not surprising considering the large negative charge of NASP and the positively charged nature of core histones.

Display large version of this figure

FIGURE 6

Binding of histones to sNASP. HeLa cell core histones H2A-H2B, H3-H4, and linker histones H1 and their trypsin-resistant core H1c were mixed with sNASP in the presence of 100mM NaCl, 10mM Tris-HCl (pH 7.5) at increasing histone:sNASP molar ratios (0 (lanes 1, 6, 11, and 16), 0.5 (lanes 2, 7, 12, and 17), 1.0 (lanes 3, 8, 13, and 18), 1.5 (lanes 4, 9, 14, and 19), and 2.0 (lanes 5, 10, 15, and 20)) and (A) analyzed by either SDS-(15%) PAGE or (B) native (6%) PAGE. C, histone H1-NASP complex; D, sNASP dimer; M, sNASP monomer; N, sNASP.

Linker histones differ from core histones in their lysine-rich composition and in that their folded protein domains are different. Core histones consist of a histone fold 56, whereas linker histones contain a winged helix motif 57. The specific binding of histone H1 to NASP raises the possibility that the winged helix domain of H1 may involved. To test this possibility, we studied the interaction of the trypsin-resistant core of HeLa cell histone H1 (which encompasses the winged helix domain). As can be seen in FIGURE 6B, this histone domain exhibits no binding to NASP, indicating that the charge effects and probably the distribution of such charge in the histone H1 molecule are responsible for the binding specificity.

To check whether sNASP facilitates the proper deposition of linker histones onto linker histone-depleted nucleosome arrays, a native chromatin fraction consisting of a weighted average nucleosome number of 23 was obtained by sucrose gradient fractionation of chicken erythrocyte chromatin 45. The NaCl dependence of the sedimentation coefficient of the fraction before and after careful depletion of linker histones (H1 and H5) was examined (FIGURE 7A). The increase in the sedimentation coefficient observed in the range of 0–80 mM NaCl measures the extent of folding of the chromatin fiber 58,59,60. Linker histones play a critical role in the folding of the chromatin fiber, and in their absence, the nucleosome array complexes exhibit a significantly lower ability to fold (open circles in FIGURE 7A). These results demonstrate the ability of NASP to add linker histones to nucleosome arrays to produce a chromatin fiber that folds in the same way as the native counterpart.

Display large version of this figure

FIGURE 7

NASP facilitates the deposition of linker histones onto nucleosome arrays. (A) Change in the sedimentation coefficient (s_20,W) of chicken erythrocyte chromatin fraction as a function of the sodium chloride concentration. The average number of nucleosomes of the chromatin fraction used was N_w=23. Data obtained before (solid circles) and after (open circles) removal of linker histones. Values obtained with the linker histone-depleted fraction on addition of histone H5 in the presence of NASP (solid triangles). (B) SDS-PAGE analysis of starting chromatin fraction (lane SS). Linker histone-depleted fraction on addition of 2mol H5/mol nucleosome in the presence (lane 1) or absence of sNASP (lane 2). Products recovered from MgCl₂ (5 mM) precipitation (lanes 3 and 4) of the samples (lanes 1 and 2, respectively). (C-E) Native-(4.5%) PAGE analysis of the DNA from the micrococcal nuclease digestion products of the starting native chromatin fraction (NC) and the linker histone-depleted counterpart on reconstitution with H5 in the presence (NASP (+)) or the absence (NASP (−)) of sNASP at 0 and 80mM NaCl in 10mM Tris-HCl, 0.1mM EDTA (pH 7.5). The triangles indicate increasing time of micrococcal nuclease digestion: 0 (lanes 1 and 6), 5 (lanes 2 and 7), 20 (lanes 3 and 8), 40 (lanes 4 and 9), and 60 (lanes 5 and 10) min.

When the linker-depleted fraction was directly mixed with histone H5 at a molar ratio of 2mol of H5/mol nucleosome in the absence of sNASP, a sedimentation coefficient of 48S was obtained in 10mM Tris-HCl (pH 7.5). This is almost identical to that of the native nonlinker depleted counterpart (result not shown). However, dialysis of this sample against 80mM NaCl resulted in its complete precipitation, indicative of an anomalous deposition. However, when the same experiment was carried out in the presence of sNASP, the values of the sedimentation coefficients at 0mM and at 80mM NaCl in the presence of 10mM Tris-HCl (pH 7.5) produced values that were slightly higher than those of the native fraction but were still very similar and followed the same increasing trend (FIGURE 7A, solid triangles). The slightly higher values are most likely the result of the native chicken chromatin fraction containing ∼1.3–1.5mol of linker histone (H1 and H5) per nucleosome 61 compared to the reconstituted complexes that can take slightly higher amounts 47 (see below).

The native chromatin fiber and its fractions are completely precipitated at MgCl₂ concentrations above 1.5mM 60,62,63. In the absence of linker histones, or when they are not properly positioned on the fiber, chromatin exhibits an enhanced MgCl₂ solubility. Therefore, such divalent ion-associated behavior can be used to monitor the proper assembly of the chromatin fiber, and it has been used to prepare highly homogeneous reconstituted oligonucleosomal fiber fragments 47,64. As can be seen in FIGURE 7B, sNASP-assisted linker histone deposition (FIGURE 7B, lanes 1 and 3) produced complexes that were precipitated in 5mM MgCl₂ and exhibited a stoichiometry of ∼1.7 mol histone H5/mol nucleosome (lane 3). In contrast, direct mixing (FIGURE 7B, lanes 2 and 4) produced complexes with an altered solubility that resulted in highly insoluble histone aggregates (lane 4). Such complexes were highly resistant to micrococcal nuclease digestion at low salt (FIGURE 7C).

The fidelity of the NASP-assisted reconstitution of chromatin can be assessed by micrococcal nuclease digestion. FIGURE 7DE, shows the digestion of complexes reconstituted in this way. Although there are slight differences in the nucleosome repeat pattern in FIGURE 7DE, between the native and NASP⁺ lanes, the pattern of nucleosome digestion is very close to that of the corresponding native counterpart at both 0 and 80mM NaCl.

Knowledge about the histone deposition and chromatin assembly mediated by histone chaperones in an in vivo setting within the cell is still very incomplete. Most of what is known comes from information on core histone chaperones that bind to either core histones H3-H4, H2A-H2B, or to both. Linker histones (histone H1 family) play a critical role in proper chromatin folding and dynamics 19 and have been shown to be deposited simultaneously with core histones on newly replicated DNA 65. Yet the mechanisms involved in histone H1 deposition and those involved in the interactions with its sole genuine chaperone (NASP) described to date for this histone family 16 remain undetermined.

We surmised that if NASP is a key player in chromatin assembly during DNA replication 22, then the protein should be ubiquitously distributed throughout the eukaryotic domain. The high cross-reactivity of NASP antibody across different vertebrate and chordate species as well as our ability in silico to identify related proteins in organisms ranging from fungi to mammals (Fig. 1) support this notion. The relatively large extent of sequence variability of NASP-like proteins observed in invertebrates (FIGURE 1A) is not surprising because these proteins correspond to chaperones for invertebrate H1 histones, which constitute the fastest evolving group of histones and exhibit a high extent of interspecies sequence variability 66,67.

In contrast to core histone chaperones such as nucleoplasmin, which has an almost exclusively β-strand secondary structure 9, or NAP-1, which consists of a mixture of α-helix and β-strand 4, the structural organization of sNASP is almost exclusively α-helical (Fig. 3). Thus, when compared to other histone chaperones, this represents a distinctive feature of this molecule. However, as with NAP-1 4,68, the protein exists in solution mainly as a dimer (Fig. 4) with a highly extended conformation. The dimeric nature does not come as a surprise because there is a predicted leucine zipper (between residues 230 and 251) in the protein sequence. An extended conformation of the homodimer would fully expose the three tetratricopeptide protein-protein interaction domains TPR1-3 14,16 that are present in the molecule (FIGURE 5C), making them available for interaction with other proteins. Indeed, several NASP nonhistone partners have been identified such as heat shock protein 90, DNA-activated protein kinase, and ATP-dependent DNA helicase II (70-kDa subunit) 14. In addition, NASP has been shown to be part of the C. elegans transformer-4-NASP– class I histone deacetylase complex, which participates in sex determination 69, and NASP is also found in the multisubunit complexes CAF-1 and histone cell cycle regulation defective homolog A that participate in histone H3.1 and H3.3 nucleosome assembly, respectively 20. The existence of so many interacting partners attests to the multiple functionality of this protein 22.

Trypsin digestion of sNASP was useful in corroborating some of the prediction analysis regarding the secondary structure organization of sNASP (FIGURE 5AB). Only in the case of one of the trypsin-resistant peptides (peptide 1, FIGURE 5BC) did the resistance extend beyond the predicted helical domains (FIGURE 5CIII). However, the protection observed in this instance spanning from residue 100 to 160 is most likely the result of the presence of glutamic acid clusters in this region corresponding to the regions 115–128 and 138–172, which had been previously identified as the putative histone H1 binding domains 16,70 and which correspond to similar glutamic acid-rich domains (108–119) and (296–326) of X. laevis N1 71 (see also FIGURE 1A). More importantly, this type of analysis helped us uncover a new glutamic acid-rich region spanning approximately from residue 280 to 300 (Fig. 5).

Our histone-binding results strongly suggest that although sNASP can bind nonspecifically to core and linker histones, it forms specific complexes only with the latter (FIGURE 6B). Linker histones are the native substrate of NASP in vivo 16,17. A similar situation has been observed with other histone chaperones with regard to histone preference. In vivo, NAP-1 and nucleoplasmin have been shown to be preferentially bound to H2A-H2B 72,73,74, whereas the Xenopus N1/N2 heterodimer and ACF-1 are found associated with H3-H4 9,11,51. However, both NAP-1 and nucleoplasmin have also been shown to be able to bind to H3-H4 9,68,75,76 and to histone H1 77,78,79 in vitro.

The ambivalent nature of all these proteins for their histone-binding substrate, especially in experiments carried out in vitro, is not surprising because regardless of their structural organization, all of them have a low pI, which undoubtedly will lead to some binding redundancy for basic proteins. The binding data also indicate that the binding of histone H1 with sNASP appears to involve the N-terminal and C-terminal domains of this histone. This is similar to what has been observed with NAP-1, where the histone tails appear to be critical for binding 75,80. The way histone tails provide binding specificity considering their disordered conformation in solution may involve the charge distribution signature conferred by their underlying primary structure. Also, they can exhibit a significant extent of secondary structure organization when interacting with DNA in the case of both linker histones 81,82 and core histones 83. Therefore, the possibility that such organization could also participate during binding to their chaperones cannot be excluded.

The binding of core histones to NAP-1 and of linker histones to NASP contrasts with that of core histones to nucleoplasmin in which the histone fold appears to be critical for binding recognition 76. This most likely reflects the differences in the diverse structural levels of folding between these chaperones to which we have previously referred. Interestingly, the three sets of domains for H1 binding to sNASP (FIGURE 5CIV) exhibit an interesting arrangement in which the two N-terminal binding sites are separated from the C-terminal domain by ∼100 amino acids. We would like to speculate here that the two distinct binding domains defined in this way serve for binding to the C-terminal and N-terminal charged regions of linker histones, with their winged fold domain occupying the region in between.

Our chromatin reconstitution experiments using a native chromatin fraction depleted of linker histones show that NASP can efficiently put the linker histones back to produce a chromatin fraction that exhibits a NaCl folding dependence almost identical to that of the native counterpart (Fig. 7). This not only attests to its physiological role in histone H1 deposition in the cellular setting 15 but also provides a nice in vitro tool that may complement the protocols already available for proper reconstitution of linker histones onto nucleosome arrays reconstituted onto sequence-defined DNA templates 47,64,84.

The results presented in this article provide strong support for the functional role(s) proposed for NASP. In vivo this molecule has been shown to transport histone H1 into nuclei 14, where it is exchanged with DNA 15. Such an exchange is likely to play a critical role in the modulation of the proper chromatin fiber folding that is critical to fundamental metabolic processes of the cell such as replication and DNA repair.