Chromatin
Structure and Superstructure
Essentally all the DNA in
eukaryotic nucleus is complexed with histone in a basic 10 nm diameter
chromatin fibre, which may be fruther folded into a 25 – 30 nm diameter fibre.
The basic fibre (or nucleous filament) is a linear array of connected nucleosomes
which about each other. Each nucleosome contains about 200 base pair (bp) of
DNA (166-241 bp depending upon source) assosaited with an octameric protein
core comprising pairs of each other of four types of histone-the lysine-rich
histone H2A and H2B, and the arginine-rich histone H3 and H4 which have been
particularly well conserved during evolution. One molecule of the fifth
histone, H1, is associated with the outside of each nucleosome, binding partly
to linker DNA between nucleosome cores.
There have been several recent
reviews of the nucleosome 53 and of chromatin more generally41,77,
so no attempt is made here to be comprehensive. Rather some aspects of
chromatin structure that are currently of particular in interest will be discussed. These include the detailed
structure that are currently of
particular interest will be discused. These include the detailed structure of
the nucleosome and its placement has any functional significance over and above
its structural significance as a parking unit. Specifically, is there ‘phasing’
of nucleosome and DNA sequences? The nature of the next level of folding of the
nucleosome filament will be discussed; the stability of such higher order
structure could in principle determine whethera particular region of chromatin
is rendered accessible for transcription, or masked and inaccessible. Finally our
current view of the structure of transcriptionally active chromatin chromatin,
recentially reviewed elsewhere41,76,124, will be summarized.
Figure 1. Structure Chromatin and Activity
Structure
of nucleosome
An outline
The nucleosome filament shows
periodic diffrential susceptibility, to several endonuclease e.g. endogenous
nuclease, micrococcal nuclease, DNase II. Evidently as the DNA leaves the
surface of one histone octamer and passes to the next it becomes more sensitive
to these enzyme than the DNA protected by hitone in the core, and nucleosomes
or oligonucleosome are released as a result of double strand cuts in these
regions of ‘linker DNA’. Mononucleosome initially contain a full repeat length
of DNA (e.g 200 bp in rat liver). Continued digestion of these particles by
micrococcal nuclease, acting as an exonuclease , reveals barriers to fruther
digestion, (presumably) imposed by histone-DNA interactions. The first such
impediment arieses when the DNA is trimmed from its full repeat length (e.g 200
bp) to 166 bp, which occurs without any change in histone composition of the
particles. A more substanstial barrier aries when the DNA is trimmed to 146 bp
(originally asigned as 140 bp), giving a particle depleted of H1, which is a
metastable intermediate in digestion and protected from fruther attack by the
octameric histone core. This paticles isolated at the two stages of digestion
have been termed chromatosome (166 bp) and nucleosome core particles (146 bp).
The digestion of the nucleosome by micrococcal nuclease can therefore be
summarized as follows :
Nucleosome
Chromatosome Nucleosome core
Particles
200 bp+
166 bp 146 bp
+ + +
octamer octamer octamer
+ + +
H1 H1
All three particles sediment about
11-12S. Analysis of their DNA sizen reveals that the broad band observed for
intact nucleosome (width ~60 bp) is sharpened during digestion so that the
nucleosome core particles DNA content is much less heterogenous (146±6 bp).
This relative homogeneity in DNA content is an important factor in archieving
the crystallization of core particles (see below). An external location of the
DNA in the nucleosome is indicated both by the accestability of DNA along its
whole length to DNase I83, and by nueron scattering studies37,86.
The nicking by DNase I at intervals of about 10 nucleotides along each strand
is taken to reflect the periodicity of B-from DNA wound around a histone core
(see also next section).
Figure 2. Histone H1
x-ray diffraction by single
crystals, combined with electron microscopy of crystal, has shown the 146 bp
core particles to be a slightly wedge-sharped disc, 11nm in diameter and 5.5 nm
high, containing about 1.75 turnx of DNA with about 80 bp per tun29.
The simplest assumption, in the absence of evidence to the contrary, is that
DNA duplex is smoothy bent around the histone core in a regular superhelix
which is left-handed32,71. The presence of a dyad axis in the
nucleosome core particle has been demonstrated uniquevocally by higher
resolution X-Ray diffraction study28, and by neutron diffractiom by
single crystals11.
Extension of the 1,75 turns if DNA
in the 146 bp nucleosome core particle by 10 bp each andwould give a particle
containing two complete turns of DNA. The corresponds to the 166 bp
chromatosome are trimmed to core particles suggests that H1 binds to the 10 bp
extensions at the ends of the core particle. The presence of H1 gives the
nucleosome filament a zigzag apperance in electron micrographs, in contrast
with a beeds-on-a-string apperance when H1 is absent, and prevents the
unfolding of nucleosomes at very low ionic strength (˂0.2 mmol/ℓ), at least on
microscope grid112. The conculsion drawn from all these observation
is that H1 seals two complete turns of DNA around the nucleosome and fixes and
the entry and exit points close together.
H1 has three distinct regions of
amino acid sequence120 which
appear to correspond to three ‘domains’ of structure, namely a central globular
region of about 80 residues, which is relativity conserved in amino acid
sequence from species to species, flanked by two very basic regions which
appear to be flexible and not tightly folded, at least in solution36.
The globular portion, isolated after tryptic removal of the flanking regions,
seems to be sufficient to restore to H1-depleted chromatin a pause in the
digestion at 166 bp3, suggesting that this domain of H1 binds to the
two terminal 10 bp regions in the chromatosome, closing two superhelical turns
of DNA. For condenstation of chromatin the basic flanking regions are required,
and these probably interact with the linker DNA, in a manner as yet unclear.
Both the nucleosome core particle
and the core histone octamer prossess a dyad axis symmetry, raising the
possibility that there may two
binding sites for H1. Although the number H1 molecules per nucleosome has been
in some dispute, recent measurements have shown that nuclei from several
sources contain only sufficent H1 for one H1 per nucleosome on average.
The Linking Number
Paradox
The paradox is that although X-ray
analysis (see above) shows that threre are two superhelical turns of DNA around
the nucleosome, the change in linking number of closed circular DNA extraced
from SV40 minichromosomes is only -1.25 per nucleosome. The change in linking
number (where the linking number is essentially the number of times one strand
of DNA duplex crosess another) aries
from some combination of change in the twist and writhe of the DNA. The number
of physical superhelical turns of DNA around the nucleosome (essentially, the
writhe) will therefore equal of change in linking number only if the helical
periodicity of the DNA (i.e the twist) is the same for DNA free in solution and
DNA associated with histone in the nucleosome. It was pointed you out29
that the paradox might be resolved if the helical periodicity was reduced from
10.5 bp per turn in solution to 10.0 bp on the nucleosome, the letter being the
value suggested by the DNase I digestion pattern (see above), other proposals
have also made for resolution paradox106,131.
DNA in solution has subsequently
been found, by two independent methods of measurement94, 122, to
have an average helical periodicity of about 10,6 bp per turn. It would
therefore seem that the linking number paradox disappears. However
redetermination of the distance between DNase I cutting sites in chromatin gave
a value of 10.4 bp on average. Thus the
peridiocity of cutting by the DNase I on the nucleosome essentially the same as
the helical peridiocity of DNA in solution, and the linking number paradox
remains. However, it has been argued48, 91 that the nuclease cutting
peridiovity of 10.4 bpon the nucleosome is compatible with a helical repeat of
10.0 bp since acces of the relatively large enzyme (DNase I) to the DNA duplex
will be restricted by the other superhelical turn of DNA, so that the enzyme does
not cut that the outherwise maximally exposed phosphodiester bonds. This
explanation would remove the linking number paradox, since the helical
peridiocity of the DNA (the twist) changes, on binding to histone octamer, from
10.6 bp to 10.0 bp per turn. The X-ray diffraction patterns from nucleosome
core particle crystals do indeed show strong intensities at 0.34 and 1.2 nm
characteristic of B-from DNA with 10.0 bp per turn.
No comments:
Post a Comment