Whole-genome chromatin profiling from limited numbers of cells using nano-ChIP-seq

Chromatin immunoprecipitation (ChIP) combined with high-throughput sequencing (ChIP-seq) has become the gold standard for whole-genome mapping of protein-DNA interactions. However, conventional ChIP protocols necessitate the use of large numbers of cells, and library preparation steps associated with current high-throughput sequencing platforms require substantial amounts of DNA; both of these factors preclude the application of ChIP-seq technology to many biologically important but rare cell types. Here we describe a nano-ChIP-seq protocol that combines a high-sensitivity small-scale ChIP assay and a tailored procedure for generating high-throughput sequencing libraries from scarce amounts of ChIP DNA. In terms of the numbers of cells required, the method provides two to three orders of magnitude of improvement over the conventional ChIP-seq method and the entire procedure can be completed within 4 d.     

Noninteger pitch and nuclease sensitivity of chromatin DNA.

Assuming that variation of nuclease sensitivity along nucleosomal DNA can basically be attributed to orientations of sugar--phosphate bonds relative to histone core, the pitch of chromatin DNA is estimated to be 10.33--10.40 base pairs. This is in accordance both with the known measured average distance between cleavage sites (10.3--10.4 base pairs) and with published data on variation of relative sensitivities of these sites to nuclease attack. The variation can be explained solely as a result of the systematic change of orientation of sugar--phosphate bonds of sensitive sites without additional suggestions about local steric hindrances by histone molecules. According to the analysis locations of sites least sensitive to nuclease attack should not depend on kind of endonuclease though the stagger could differ. We conclude that the nucleosome core particle is axially symmetrical. The results strongly support the suggestion that DNA is wrapped around the histone octamer smoothly, without interruption of base-stacking interactions.     

Differential nuclease sensitivity of the ovalbumin and beta-globin chromatin regions in erythrocytes and oviduct cells of laying hen.

We have monitored the differential nuclease sensitivity of defined regions of the chicken genome in different cells using a method which combines restriction enzyme digestion and blotting to diazobenzyloxymethyl (DBM)-paper (see Ref. 11). By using different specific probes and by scanning the bands on the autoradiograms, it is possible to compare on the same blot the digestion patterns of similar-sized fragments from different regions of the genome corresponding to "active" and reference "inactive" genes. We have demonstrated the preferential sensitivity to DNaseI and micrococcal nuclease digestion of the ovalbumin gene region in hen oviduct chromatin. The beta-globin gene region (containing both an adult and an embryonic gene) is also preferentially digested by DNaseI in hen mature erythrocyte nuclei, but at a lower rate than the ovalbumin gene region in oviduct. These observations raise the possibility that there may be several types of preferential nuclease sensitivities, all characterized by increased rates of digestion but to different levels, the highest corresponding to the very actively transcribing genes.     

Internucleosome interaction: detection of dinucleosome fragmentation of chromatin by micrococcal nuclease. Analysis of the products of cleavage of chromatin from rat liver nuclei and L cells by micrococcal nuclease

In murine L-cell nuclei micrococcal nuclease causes chromatin fragmentation with predominant liberation of dinucleosomes. Analysis of dynamics of rat liver nuclear chromatin cleavage by micrococcal nuclease revealed that the "dinucleosomal" mode of fragmentation is due to the pretreatment of nuclei with the non-ionic detergent Triton X-100 in the course of the isolation procedure. The set of particles detected in nuclease hydrolysates of nuclear chromatin pretreated with Triton X-100 and those isolated by the standard procedure was shown to be significantly different. In Triton X-100 treated nuclei the dichromatosome is the main hydrolysate component under various experimental conditions of nuclease hydrolysis and the sole component under "mild" conditions, whereas sucrose-treated nuclei contain three types of dinucleosomes. In Triton-treated nuclei prolongation of hydrolysis results in the liberation of the chromatosome which is absent in chromatin hydrolysates of sucrose-treated nuclei. Hydrolysis of Triton-treated nuclear chromatin by micrococcal nuclease is unaccompanied by the liberation (up to the stage of "deep" hydrolysis) of the core particle, the major component of the "sucrose" nuclear hydrolysate under the conditions used. The sharp differences in the accessibility of various types of dinucleosomes observed during pretreatment of nuclei with Triton X-100 are interpreted in terms of the localization of histone H1. The non-random type of the histone H1 molecule orientation along the nucleosome fibril is postulated.     

Structural differences in the chromatin from compartmentalized cells of the sea urchin embryo: differential nuclease accessibility of micromere chromatin.

The chromatin structure of three cell types isolated from the 16-cell stage sea urchin embryo has been probed with micrococcal nuclease. In micromeres, the four small cells at the vegetal pole, the chromatin is found to be considerably more resistant to degradation by micrococcal nuclease than chromatin in the larger mesomere and macromere cells which undergo more cellular divisions and are committed to different developmental fates. The micromeres show an order of magnitude decrease in the initial digestion rate and a limit digest value which is one third that of the larger blastomeres; both observations are suggestive of the formation of a more condensed chromatin structure during the process of commitment, or as the rate of cell division decreases. The decreased sensitivity to nuclease for micromeres is similar to results reported for sperm and larval stages of development.     

Expansion of chicken erythrocyte nuclei upon limited micrococcal nuclease digestion. Correlation with higher order chromatin structure

A sensitive method for measuring nuclear volumes with a Coulter counter is described. It has been applied to the digestion of chicken erythrocyte nuclei by micrococcal nuclease and DNase I. Early in digestion, micrococcal nuclease induced a 20% increase in the effective spherical volume of the nuclei, followed by a gradual reduction. At the peak of nuclear swelling, about 17% of the chromatin was soluble after lysis and its average chain length was about 18 kilobase pairs (kb). DNase I digestion did not give rise to a corresponding expansion of the nuclei. Several preparation conditions, including the treatment of nuclei with 0.2% Triton X-100, led to a loss of the expansion effect upon subsequent micrococcal nuclease digestion. The results support the domain theory of higher order chromatin structure. In the context of this model, the observed maximum nuclear expansion correlates with an average of one nuclease scission per domain.     

The effect of histone hyperacetylation on the nuclease sensitivity and the solubility of chromatin

We have examined the effects of histone hyperacetylation upon nuclease digestion of nuclei and subsequent fractionation of chromosomal material in the presence of MgCl2. DNase I shows a maximum sensitivity towards hyperacetylated nuclei at somewhat elevated ionic strengths (150-200 mM NaCl), whereas micrococcal nuclease exhibits no specificity for acetylated nuclei over a broad range of ionic strengths. Fractionation in the presence of MgCl2 of hyperacetylated nuclei digested with micrococcal nuclease results in a substantial increase in the amount of soluble chromatin relative to that obtained with control nuclei. This increased yield of Mg2+-soluble chromatin results from the recruitment into this fraction of oligonucleosomes containing extremely hyperacetylated histones. These results suggest that contiguous nucleosomes containing highly acetylated histones may be altered in their ability to interact with themselves and with other nucleosomes.     

Digestion by micrococcal nuclease of mouse submandibular-salivary-gland chromatin.

Nucleic prepared from mouse submandibular salivary gland show marked fragility during isolation. Hwever, intact nuclei relatively free from cytoplasmic contamination were obtained by homogenization in buffers containing 0.88 M-sucrose, Ca2+, spermine, spermidine and the proteinase inhibitor aprotinin, followed by centrifugation through 2.2 M-sucrose. The kinetics of digestion by the micrococcal nuclease of chromatin in these nuclei are similar to those of chromatin from mouse liver nuclei. Base-pair size analysis of the solubilized DNA from both organs shows a stable high-molecular weight species of chromatin, which is further digested to mononucleosome and subnucleosome species. With extensive digestion the chromatin becomes insoluble. The mononucleosomes produced from salivary-gland chromatin after the inhibition of endogenous proteinase activity exhibit an s20,w value of 11S and contain histones H1, H2A, H2B, H3 and H4.     

The nuclease sensitivity of active genes.

Brief micrococcal nuclease digestion of chick embryonic red blood cells results in preferential excision and solubilization of monomer nucleosomes associated with beta-globin sequences and also 5'-sequences flanking the beta-globin gene. Both regions are DNAse-I sensitive in nuclei. Such salt-soluble nucleosomes are enriched in all four major HMG proteins but HMG1 and 2 are only weakly associated. These nucleosomes appear to have lost much of the DNAse-I sensitivity of active genes. The HMG14 and 17-containing salt-soluble nucleosomes separated by electrophoresis are not DNAse-I sensitive and contain inactive gene sequences as well as active sequences. Reconstitution of HMG proteins onto bulk nucleosomes or chromatin failed to reveal an HMG-dependent sensitivity of active genes as assayed by dot-blot hybridization and it was found that the DNAse-I sensitivity of ASV proviral sequences as assayed by dot-blot hybridization was not HMG-dependent. These results indicate that higher order chromatin structures might be responsible for nuclease sensitivity of active genes.     

Nucleosome packing in chromatin as revealed by nuclease digestion.

Chromatin DNA of rat thymus nuclei was cleaved by Serratia marcescens endonculease. The fragments have been examined by polyacrylamide gel electrophoresis under denaturing conditions. The results obtained are interpreted to mean that the internucleosomal DNA is cleaved by the endonuclease into fragments which are multiples of 10 nucleotides. The 10 nucleotide periodicity in fragmentation of internucleosomal DNA is independent of the presence of histone H1 and is likely to be determined by the interaction of this DNA stretch with the histone core of nucleosomes. Such interaction implies a close association between the nucleosomes in the chromatin thread. Quasi-limit chromatin digest (50--55% of DNA hydrolysis) contains undegraded DNA fragments with length of up to 1000 nucleotides or more. A part of this resistant DNA consists of single-stranded fragments or contains single stranded regions. These data may be accounted for by a very compact nucleosome packing in the resistant chromatin in which one of the DNA stands is more accessible to the endonuclease action.     

Effects of thyrotropin on thyroid chromatin: Enhanced sensitivity to micrococcal nuclease and increased nuclear protein phosphorylation

Thyroid slices were incubated with or without TSH for 2 or 5 h. Nuclei were then prepared, subjected to mild digestion with micrococcal nuclease, and centrifuged at 1200 × g. The amount of DNA in 1200 × g supernatants was increased by TSH at 5 h, but not at 2 h. In parallel studies, thyroid slices were incubated with ³²P_i and labeling of acid-soluble nuclear proteins was examined. TSH-dependent increases in labeling of histones H1 and H3, and of the high mobility group protein HMG 14, were observed at 2 h; however, there were no apparent changes in TSH-dependent labeling between 2 and 5 h, in nuclease-sensitive or in bulk chromatin. These results suggest that the observed TSH-dependent changes in the micrococcal nuclease-sensitivity of thyroid nuclear chromatin were not induced directly by changes in the phosphorylation of the histones or HMG 14.     

P1 nuclease defines a subpopulation of active SV40 chromatin--a new nuclease hypersensitivity assay.

Under exhaustive digestion conditions P1 nuclease was found to cleave a subpopulation of intracellular SV40 chromatin only once. The major P1 cleavage site in SV40 DNA was mapped at the origin of DNA replication, and the two minor sites at the SV40 enhancers. The P1-sensitive SV40 chromatin subpopulation was found to have higher superhelical density than the bulk of the intracellular SV40 chromatin. Furthermore, pulse labeled SV40 DNA which had higher superhelical density than that of the steady state viral DNA (S.S.Chen and M.T.Hsu, J.Virol 51:14-19, 1984) was also found to be preferentially cleaved by P1 nuclease. These results are consistent with a supercoil-dependent alteration of chromatin conformation near the regulatory region of the viral genome that can be recognized by P1 nuclease. Since P1 nuclease cleaves the subpopulation of SV40 chromatin only once without further degradation, this nuclease can be used as a general tool to define viral or cellular chromatin fraction with altered chromatin conformation and to map nuclease hypersensitive sites. Preliminary studies indicate that P1 makes limited double stranded cleavages in cellular chromatin to generate large DNA fragments.     

Cleavage of DNA in nuclei and chromatin with staphylococcal nuclease.

Treatment of either rat liver chromatin or intact nuclei with the enzyme staphylococcal nuclease results in the conversion of about half of the DNA to acid-soluble oligonucleotides. As previously described, mild digestion of nuclei results in the liberation of a series of nucleoprotein particles containing DNA fragments which are all integral multiples of a unit length DNA 185 base pairs in length. Analysis of the kinetics of appearance of these fragments suggests that at least 85% of the nuclear DNA is involved in the formation of the repeating subunit profile. More extensive digestion of nuclei however results in the generation of a series of eight unique DNA fragments containing 160 to 50 base pairs. The series of smaller molecular weight DNA is virtually identical with the profile obtained upon limit digestion of isolated chromatin. By velocity centrifugation we have obtained highly purified preparations of the monomeric nucleoprotein particle. Digestion of this monomeric subunit results in the solubilization of 46% of the DNA and analysis of the resistant DNA again reveals the set of eight lower molecular weight fragments. These data suggest that the initial site of nuclease cleavage in chromatin resides within the DNA bridging the repeating monomeric subunits. Further attack results in cleavage at a set of sites within the monomer liberating a pattern of smaller DNA fragments which probably represents the points of intimate contact between the histones and DNA.     

Response of isolated nuclei to phospholipid vesicles: analysis of chromatin sensitivity to DNase I and micrococcal nuclease

Phosphatidylserine (PS) and phosphatidylcholine (PC) multilamellar vesicles (MLV) affect chromatin structure as analysed by DNase I sensitivity. The kinetics of DNA solubilisation during the digestion of nuclei indicates that phosphatidylserine causes an increase in DNase accessibility while phosphatidylcholine slightly reduces this accessibility. The effect of phosphatidylserine has also been analysed by means of isokinetic sucrose gradients and agarose gel electrophoresis of nuclear DNA solubilised by micrococcal nuclease. This analysis indicates that phosphatidylserine induces a very rapid production of mononucleosome subunits as compared with untreated nuclei.     

S1 nuclease sensitivity of a double-stranded telomeric DNA sequence.

We examined structural properties of poly d(C4A2).d(T2G4), the telomeric DNA sequence of the ciliated protozoan Tetrahymena. Under conditions of high negative supercoiling, poly d(C4A2).d(T2G4) inserted in a circular plasmid vector was preferentially sensitive to digestion with S1 nuclease. Only the C4A2 strand was sensitive to first-strand S1 cutting, with a markedly skewed pattern of hypersensitive sites in tracts of either 46 or 7 tandem repeats. Linear poly d(C4A2).(T2G4) showed no preferential S1 sensitivity, no circular dichroism spectra indicative of a Z-DNA conformation, no unusual Tm, and no unusual migration in polyacrylamide gel electrophoresis. The S1 nuclease sensitivity properties are consistent with a model proposed previously for supercoiled poly d(CT).d(AG) (Pulleyblank et al., Cell 42:271-280, 1985), consisting of a double-stranded, protonated, right-handed underwound helix. We propose that this structure is shared by related telomeric sequences and may play a role in their biological recognition.     

Heterogeneity of chromatin fragments produced by micrococcal nuclease action.

Digestion of calf thymus chromatin with micrococcal nuclease produces a mixture of apparently well defined nucleoprotein fragments which have been partially resolved by sedimentation on linear (5-20%) sucrose gradients. Sedimentation patterns reveal a predominant peak at the 11S position, three slower components, which have not previously been reported, at the 3.4S, 5.3S and 8.6S positions, and three faster components at the 17S, 22S and 26S positions. DNA isolated from the 3S to 12S region of gradients has been resolved on polyacrylamide gels into nine to ten discrete components ranging from 47 to 156 base pairs in length. A nearly identical pattern of small DNA products was obtained from chromatin digested in intact nuclei. These data suggest that chromatin contains either several types of subunits or predominently a single type of subunit which can be asymmetrically cleaved at any one of four or more sites.     

Nuclease sensitivity of active chromatin.

The active regions of chicken erythrocyte nuclei were labeled using the standard DNase I directed nick translation reaction. These nuclei were then used to study the characteristics and, in particular, the nuclease sensitivity of active genes. Although DNase I specifically attacks active genes, micrococcal nuclease solubilizes these regions to about the same degree as the total DNA. On the other hand micrococcal nuclease does selectively cut the internucleosomal regions of active genes resulting in the appearance of mononucleosomal fraction which is enriched in active gene DNA. A small percentage of the active chromatin is also released from the nucleus by low speed centrifugation following micrococcal nuclease treatment. The factors which make active genes sensitive to DNase I were shown to reside on individual nucleosomes from these regions. This was established by showing that isolated active mononucleosomes were preferentially sensitive to DNase I digestion. Although the high mobility group proteins are essential for the maintenance of DNase I sensitivity in active regions, these proteins are not necessary for the formation of the conformation which makes these genes preferentially accessible to micrococcal nuclease. The techniques employed in this paper enable one to study the chromatin structure of the entire population of actively expressed genes. Previous studies have elucidated the structure of a few special highly prevalent genes such as ovalbumin and hemoglobin. The results of this paper show that this special conformation is a general feature of all active genes irregardless of the extent of expression.     

Protease sensitivity and nuclease resistance of the scrapie agent propagated in vitro in neuroblastoma cells.

The scrapie agent has been propagated in vitro in mouse neuroblastoma cells. To further characterize the tissue culture-derived scrapie agent, we studied the effects of protease and nuclease digestion on the agent derived from these cells. The scrapie agent in these cells was found to be resistant to protease digestions for short times but was inactivated by prolonged digestion at high protease concentrations. In contrast, digestion with a variety of nucleases did not alter the agent titer. These results demonstrate that the agent requires an essential protein or proteins for infectivity. If the agent also contains a nucleic acid genome, it must be more nuclease resistant than the majority of cellular DNA and RNA. These properties of the tissue culture-derived scrapie agent were identical to those of brain-derived scrapie agent and thus cannot be attributed to secondary effects of tissue pathology, since the infected cell cultures show no cytopathic effects as a result of infection.