Last time, we looked at all the organizational features necessary to handle gene expression in chromatin domains. Today, we need to look at the higher-level structural features required for transcription, replication and chromosome segregation.
Decondensed chromosomes in the interphase
nucleus occupy discrete territories Chromosome painting demonstrates that chromosomes occupy discrete volumes within the interphase nucleus. There are referred to as "chromosome territories". Note that the image at right is based on a 2-dimensional slice through a 3-dimensional nucleus. Not all copies of all chromosomes can be seen in any given plane. Top: FISH (Fluorescence in situ hybridization) labeling of all 24 different human chromosomes (1 - 22, X, and Y) in a fibroblast nucleus, each with a different combination of in total seven fluorochromes. Shown is a mid-plane of a deconvoluted image stack which was recorded by wide-field microscopy. Bottom: False color representation of all chromosome territories visible in this mid-section after computer classification. Andreas Bolzer, Gregor Kreth, Irina Solovei, Daniela Koehler, Kaan Saracoglu, Christine Fauth, Stefan Müller, Roland Eils, Christoph Cremer, Michael R. Speicher, Thomas Cremer - Bolzer et al., (2005) Three-Dimensional Maps of All Chromosomes in Human Male Fibroblast Nuclei and Prometaphase Rosettes. PLoS Biol 3(5): e157 DOI: 10.1371/journal.pbio.0030157, part of Figure 1. |
Researchers in Dr. John Sedat's lab at UCSF wanted to determine whether specific loci for a given homologous pair of chromosomes were physically associated during interphase. This required an imaginative approach. Yeast was transformed with the E. coli lac operator sequence, which is specifically bound by the lac repressor protein. Next, the same yeast strain was transformed with a chimeric gene in which the Green Fluroescent Protein (GFP) was added to the amino terminus of the E. coli lac repressor gene. The chimeric gene was inserted at random into a different chromosomal site. Since the chimeric lac repressor/GFP protein will only bind to the lac repressor sequence, a diploid yeast nucleus should only have two binding sites, one for each homologous locus. |
|
The time-lapse video by Wallace Marshall
shows the results. The chimeric GFP bound to the lac repressor binding site in yeast chromosomes shows up as fluorsecent signal. These results demonstrate that both homologous loci, although free to move within the nucleus, are constrained in their motion, and maintain a close association even during interphase. |
|
Collings DA et al. (2000) Plant nuclei can contain extensive grooves and invaginations. Plant Cell 12: 2425-2439.
Although we often visualize the nucleus as a sort of basketball,
studies in onion epidermal cells have shown that these nuclei are
lens-shaped, with a surface formed by grooves, invaginations and
channels. Grooves were seen to go as deep as 6 µm, and
invaginations as deep as 8 µm into the nucleus.
A - Serial sections through onion
epidermal nuclei, in which DNA has been labeled using DAPI
dye. DNA is visualized by UV fluorescence. B. - Light microscopic images of sections shown in A. C - 3D reconstruction of surface based on DAPI fluorescence. Nc = nucleoli; G = groove; arrow shows an invagination. |
from Figure 1. Copyright © 2000 by the American Society of Plant Biologists |
Recalling that the nucleus is thought to be a specialized
structure formed from the endoplasmic reticulum, what is the
importance of these findings? That is, how might a lens shape with
a convoluted surface function more efficiently than a rigid
spherical shape?
A sphere is the geometrical shape that minimizes surface area per unit volume. Both the lens shape and the channels, invaginations and groves would contribute to increasing the surface area. Since the nuclear envelope controls traffic of macromolecules into and out of the cell, that traffic can be carried more efficiently with a large surface area. In particular, grooves and invaginations extending deep into the nucleus ensure that no part of the nucleus is far from the nuclear membrane. This will minimize the time required for a transcript to exit the nucleus, or for a protein to enter, and find its way to a chromosome.
We have been discussing domains of chromatin as functional units
in gene expression, some up to 150kb in length. But we haven't yet
addressed the question, "does a chromosome have many molecules of
DNA, or only one?" There are three lines of evidence which
demonstrate that eukaryotic chromosomes are single linear DNA
molecules.
First, we have to review how semi-conservative DNA replication
works.
EXPERIMENT: DEMONSTRATION OF SEMI-CONSERVATIVE REPLICATION
When chromosomes
from the semi-conservative replication experiment are
viewed in fluroescence microscopy, we seethat
that each chromatid, and hence the chromosome, is a single
DNA molecule. If this were not so, then we would
simply see an even distribution of dye in both
chromatids, and a gradual dilution of BUdr (dark) in
subsequent cell generations. Instead, we see a discrete
partitioning of the BUdr into one or the other of
the sister chromatids, which is consistent with the idea
that each chromatid contains one old strand and one
newly-replicated strand. In some chromosomes, dye has been partitioned completely to one chromatid or the other (circled). In other chromosomes, sister-chromatid exchange is evidenced by a checkered-pattern, in which dye abruptly shifts from one chromatid to the other (arrow) image from Kimball's Biology Pages http://users.rcn.com/jkimball.ma.ultranet/BiologyPages/H/Harlequin.html |
Yeast
(Saccharomyces cerevisiae ) cells were embedded
in agarose, and the agarose plugs treated with enzymes
to degrade the cell walls. Agarose plugs are then loaded
into the wells of a pulsed-field gel. All 15 yeast chromosomes can be resolved on this gel. |
Image displayed from http://bio3400.nicerweb.com/Locked/media/ch19/pulsed-field-gel-electrophoresis.html |
ii) Linear molecules correspond in length to estimated lengths of chromosomes.
Yeast was the first eukaryotic genome to be completely sequenced. The first complete yeast chromosomal sequence was from chromosome III.
Database entry: 315339bp
182 ORFs > 100bp (mostly unknown
genes)
The complete yeast genome can be searched and browsed at the Saccharomyces Genome Database (SGD) at Stanford: http://www.yeastgenome.org/
Alternative site: NCBIAt
51000X,
it is possible to identify rounded projections of about
1.5-2x the diameter of chromatin. These have been
interpreted as the loop domains described above.
Although the resolution of SEM is good enough, sample
preparation technology has not yet made it possible to
observe finer structure in chromatin loops. It is also
worth pointing out that in the figure we are actually
seeing two sister chromatids, which will separate
during anaphase. This is evidenced by the gap between the
chromatids. Note the interchromatid fibers, which are
about the right size for individual chromatin solenoids. (Allen et al.,Fig. 4 p57) Detail of a single chromosome viewed at high resolution in the SEM. The majority of the chromosome surface displays a twisted loop configuration with approximately twice the diameter of the individual fibers which are seen mainly as interchromatid fibers (arrowed). (Magnification x 51000). |
Now
let's
zoom out by a factor of 5 to look at complete chromosomes.
(Allen et al. Fig. 3, p56) FIG. 3. Part of a metaphase spread preparation prepared without banding techniques, illustrating small circumferential grooves on the surface of the chromatids. (Magnification x 9500). Some segmentation and indentations of chromosomes apparent, but now compare with scanning EM of G-banded chromosomes. |
Here,
we
can see 'circumferential grooves' that run as a
helix down each chromatid. Again, this indicates helical
organization at the highest level of chromosome
structure. It's important to realize that when we do
any kind of banding technique, we are altering the
chromosome in some way,such that they take up dye
more readily. Usually, a partial hydrolysis of protein
with the protease trypsin is used. So you have to remember
that what we're seeing is not the way the untreated
chromosome would actually look, if we could see it with
light, but rather an altered structure whose purpose is
to bring out structural features. |
(Allen et al., Fig. 5, p58) Low-power micrograph of a G-banded metaphase spread. The chromatid arms are segmented by circumferential grooves which run as a helix down each chromatid. (Magnification x 2800). |
Two alternative models of
high-level chromatin folding during prophase are presented
in A.
Figure
9. Licensed under Creative Commons
Attribution-Noncommercial-Share Alike 3.0 Unported
license, as described at http://creativecommons.org/licenses/by-nc-sa/3.0/
and http://creativecommons.org/licenses/by-nc-sa/3.0/legalcode.
|
To test these hypotheses,
constructs were made containing 8 tandem repeats of the E.
coli lac operator sequence, flanked by SAR sequences. This
construct was transfected into Chinese Hamster Ovary (CHO)
cells in culture. Transformed cells, which had
incorporated the construct into chromosomes, were screened
by flow cytometry for cells that contained large numbers
of copies of the construct. One cell line, dSAR-d11, which
had approx. 1000 tandem copies of the construct inserted
at a single locus. |
The images in Fig. 9B are from chromosomes incubated with gold-tagged antibodies to the lac repressor protein. This immuno-gold staining method detects lac repressor binding to the chromosomes as bright gold particles. Fig. 9B a-c are examples of slices visualized by electron microscopy, in X, Y and Z axes. A stack or serial images, spanning the thickness of each chromosome visualized, can be deconvoluted in software to create a 3D-image of the chromosome, showing the location of the lac operator sequence in the chromosome. |
Fig. 9C is a
3D-reconstruction of one chromosome.
|
The highest levels of coiling are probably mostly absent in the
interphase nucleus. However, at any given time, different regions
of chromosomes may be coiled or uncoiled, to one degree or
another. Highly-coiled chromosomal regions will not be genetically
active.
Electron Microscopy Tomography (EMT) of DNA-Depleted Human (HeLa) Cell-Line Chromosome Human chromosomes were treated with DNAseI to remove most of the DNA, and slides were scanned in layers by electron microscopy. Layers were reconstructed into a 3-dimensional image. |