Differentiation and gene regulation
Editorial overview
Frank G Grosveld and Denis Duboule
Current Opinion in Genetics & Development 2007,
17:369–372
0959-437X/$ – see front matter
# 2007 Elsevier Ltd. All rights reserved.
DOI 10.1016/j.gde.2007.10.001
Frank G Grosveld
Department of Cell Biology, Erasmus Medical
Centre, Rotterdam, The Netherlands
e-mail: f.grosveld@erasmusmc.nl
Frank Grosveld is a professor of Cell
Biology at the Erasmus Medical
Center in Rotterdam. He is a
biochemist and molecular biologist by
training and is active in the field of
gene regulation in mammalian
development and disease. He is a
member of two EU networks of
excellence ‘‘the Epigenome’’ and
‘‘Cells into Organs’’ and the
worldwide consortium ‘‘the
International Regulome’’ that aims to
characterise all transcription factor
complexes, and which includes
coordination of the EU integrated
project ‘‘EUTRACC’’.
Denis Duboule
National Research Centre ‘Frontiers in
Genetics’, Department of Zoology and Animal
Biology, University of Geneva, Sciences III and
School of Life Sciences, Federal Institute of
Technology, Lausanne, Switzerland
e-mail: denis.duboule@zoo.unige.ch
Denis Duboule is professor of
Developmental genetics and
genomics at the University of Geneva
and the Federal Institute of
Technology in Lausanne. He is the
chair of the Swiss National Centre
‘Frontiers in Genetics’ and is active in
the fields of developmental genetics
of vertebrates, in particular
concerning large scale gene
regulation, in an evolutionnary
context.
www.sciencedirect.com
Introduction
Over the past 30 years, our views on the mechanisms underlying gene
regulation have changed considerably. It is now well accepted that, in order
for a transcription unit to be properly regulated, not only should the
requested DNA sequence-specific factors and complexes thereof be pre-
sent, but also the micro-environmental conditions, in terms of local chro-
matin configuration, should be appropriate. The global organization of
chromosomes and the dynamics of the genome in the nucleus are equally
important and have been a focus of attention since the beginning of the 20th
century (e.g. Wilson 1905 [4]). Although a lot of progress on the basic
principles and processes was made, it remained very difficult to study the
interactions that take place in the genome. However, with the advance of
new technologies and a spectacular increase in the size of the toolbox
available to present day investigators, it has only fairly recently become
possible to study the nucleus at the level of resolution required to measure or
visualize interactions and the movement of individual gene loci or whole
chromosomes. As a result recent developments have been fast and furious,
covering a range of interests from studies on the role of single proteins to the
analysis of the interactions between complete chromosomes.
This fast pace inevitably leads to contradictory observations and it is
sometimes difficult to distinguish fact from fiction, particularly when novel
technologies are used or when techniques depend on the eye of the
investigator, as is the case in a number of microscopic techniques. Never-
theless, the study of nuclear organization and dynamics is an interesting and
exciting area of research aiming to understand how the cell ‘organizes’ a
large genome into the small space of the nucleus; which mechanisms it uses
to regulate the expression of the genome while maintaining the flexibility to
respond to cues coming from other cells or from the environment; and how
the cell duplicates and divides the genome into two daughter genomes. One
of the most debated issues in the field is whether nuclear structure adheres
to some ‘masterplan’ or ‘design’ or whether its organization is based on
stochastic principles and dependent on the remarkable properties of macro-
molecules and macromolecular complexes.
The reviews in this issue of Current Opinion discuss novel concepts and
technologies associated with the control of gene expression during de-
velopment, as viewed from different angles. Topics range from examining
basic physical principals relevant to nuclear organization to the review of
proteins involved in genomic interactions and the (dynamics of) interactions
within a chromosome and between chromosomes.
Organizing principles and genome variation
The nucleus is organized in sub-compartments with distinct biological
activities, which represents an important regulatory layer for cell function.
Current Opinion in Genetics & Development 2007, 17:369–372
370 Differentiation and gene regulation
Karsten Rippe discusses new insights into the principles
by which nuclear organelles form. This usually occurs in a
self-organising manner based on the specific physical
properties of macromolecular complexes, leading to the
formation of stable but plastic structures involving
multiple, yet relatively weak, interactions. Changes in
the nuclear environment or, alternatively, addition,
deletion or modification of individual components can
rearrange such structures leading to a different functional
state. Importantly, it describes how movement within the
nucleus or even at the level of a chromosome can take
place without the involvement of a directed force such as
for example provided by motor proteins.
The genome itself is not a fixed entity with small vari-
ations and recent work has shown that large stretches of
the genome can vary considerably. Alexandre Reymond,
Charlotte N. Henrichsen, Louise Harewood and Giu-
seppe Merla review the first extensive catalogue of struc-
tural human variations that was recently released. It
showed that copy number variation in large stretches of
the genome is extremely abundant, raising the possibility
that they play a major role in functional variation. Geno-
mic insertions and deletions obviously contribute to phe-
notypic differences by modifying the expression levels of
genes within the aneuploid segments. Interestingly, how-
ever, such variations also influence the expression of
neighbouring genes present in normal copy numbers
and the authors discuss the possible mechanisms behind
this latter effect. Structural properties of the genome can
also be altered by modifications of the chromatin. Anton
Wutz and Joost Gribnau discuss the most dramatic
example of epigenetic modification of the genome, X
inactivation, which in mammals leads to an almost sep-
arate domain within the nucleus. Inactivation of one of
the two female X chromosomes is essential to establish
dosage compensation between XY males and XX females
in placental mammals. The choice of X inactivation is
random and is controlled by the X inactivation center
(Xic). Recent advances in genome sequencing show that
the Xic has evolved from an ancestral vertebrate gene
cluster in placental mammals and underwent separate
rearrangements in marsupials. The Xic ensures that all
but one X chromosome per diploid genome are inacti-
vated. Pairing of Xic loci on the two X chromosomes and
alternate states of the X chromosomes before inactivation
have recently been implicated in the mechanism of
random choice. Chromosome-wide silencing is then
initiated by the non-coding Xist RNA, which evolved
with the mammalian Xic and covers the inactive X
chromosome.
Interactions within chromosomes loops and
boundaries
The development of a number of technologies but in
particular the Chromosome Conformation Capture (3C)
approach (Dekker et al. [2]) adapted to the analysis of
Current Opinion in Genetics & Development 2007, 17:369–372
individual loci (Tolhuis et al. [3]) has resulted in a large
leap forward in our understanding of how loci are orga-
nized in loops. Genetic studies in Drosophila spearheaded
the initial ideas and understanding of how loci may be
organized in such loops between boundaries and Robert,
K. Maeda and François Karch review how regulatory
elements can exert their effect over many tens of kilo-
bases of DNA on particular promoters, using chromatin
boundaries in the fruit fly. The existence of boundaries
was first proposed on the basis of Drosophila genetics.
Because these elements are involved in such diverse
processes and show little or no sequence homology be-
tween each other, no single molecular mechanism could
account so far for their activity.
Recent evidence nonetheless showed that boundaries
probably function through the formation of long-distance
chromatin loops. These loops have been proposed to play
a crucial role in both controlling enhancer–promoter
interactions and packing DNA. The molecular mechan-
ism and some of the molecules involved in loop and
boundary formation are discussed by Julie A. Wallace
and Gary Felsenfeld, with a special emphasis on two
different ‘organizer’ proteins. They review the best-
characterized elements involved in the separation of
functional domains, the gypsy in Drosophila and the
CTCF-binding element in vertebrates. These sequences
stabilize contacts between distant genomic regulatory
sites leading to the formation of looped domains. They
discuss CTCF mediating contacts in the mouse b-globin
locus and, at the Igf2/H19 imprinted locus, the formation
of active chromatin hubs and transcription factories. The
properties of CTCF, most of which is stably bound to the
chromatin, and its recently described genomic distri-
bution, suggest that it plays an important role in nuclear
architecture. Sanjeev Galande, Prabhat Kumar Purbey,
Dimple Notani, and Pavan Kumar describe the role of
another nuclear ‘organiser’ protein, SATB1. They discuss
the function of SatB1 as a key factor for the integration of
higher-order chromatin architecture with gene regulation.
SATB1 organizes the MHC class-I locus into distinct
chromatin loops and plays a key role in the response to
physiological stimuli. At a genome-wide level, SATB1
seems to act as an organizer of transcriptionally poised
chromatin.
Regulation within large domains
The structure/function organization of some large loci,
which heavily depend on looping factors, is discussed
next. Cornelis Murre reviews the control of immunoglo-
bulin gene rearrangement. This locus extends over more
than 1.5 Mb and is of interest from an organizational point
of view. At early stages of B cell development, one of the
variable regions is recombined with the D, J and constant
regions in a process known as VDJ recombination. A key
aspect of this process is how distant V regions can recom-
bine with efficiencies similar to those seen for the V
www.sciencedirect.com

regions that are located much closer to D and J. Much is
known about the control of its transcription and epige-
netic remodelling but it is only recently that data are
emerging which suggest that there is an underlying
structural order that facilitates the association of DNA
elements separated by large genomic distances.
Another well-known example of the importance of gene
proximity for shared and coherent transcriptional regula-
tion is provided by the Hox gene family. In vertebrates,
these genes are organized in tight genomic clusters and
instruct early embryonic tissues about their positional
identities in a process called patterning. Jacqueline
Deschamps discusses the relationship between clustering
and gene regulation. Hox gene clustering is one of the
most interesting aspects of their functioning because the
order of the genes in each cluster largely reflects the
anterior to posterior order of their expression in the early
embryo. This link between gene order and order of
expression, termed collinearity, originated early during
evolution and has been conserved from flies to human.
Deschamps reviews how murine Hox genes are regulated
in part by gene-proximal regulatory elements, but inter-
estingly how several of their essential spatial expression
properties are dependent on global regulatory elements
located outside the complexes, leading to the notion of
‘regulatory landscape’.
Another regulatory landscape crucial for proper vertebrate
development is reviewed by Rolf Zeller and Aimée
Zuniga, who discuss the process of patterning from a
different perspective, that of limb development. During
this process, two regulatory signals play essential roles in
digit patterning, the Shh morphogen and the BMP
antagonist Gremlin1. Their expression patterns are very
dynamic and regulated by cis-regulatory elements
embedded in unrelated and remote neighbouring genes.
Malfunction of elements that can be located at megabase
distances from the gene are the primary cause of different
types of congenital limb malformations. Recent compara-
tive and functional genomics studies have uncovered
large and complex chromosomal landscapes controlling
these genes, in which some HOX products play an
important role.
Chromatin mobility
One of the most interesting, yet hotly debated, issue
relates to the mobility of the genome and is reviewed
by Evi Soutoglou and Tom Misteli. They describe how
chromatin is increasingly recognized as a highly dynamic
entity. Chromosome sites, in both lower and higher
eukaryotes, undergo frequent, rapid and constrained local
motion and occasional slow, long-range movements.
While the dynamic properties of chromatin have been
described by visualization in vivo, some of the functional
relevance of chromatin mobility has only recently become
clear. Both the mobility and immobility of chromatin
www.sciencedirect.com
Editorial overview Grosveld and Duboule 371
appears to have functional consequences: local (diffu-
sion-based) motion of chromatin is important in gene
regulation, yet global chromatin immobility appears to
play a key role in maintenance of genomic stability.
Interactions between chromosomes
The interactions between chromosomes and between
chromosomes and the nuclear envelop have recently
become a focus of intensive studies. Giacomo Cavalli
reviews how chromosomes occupy distinct territories in
the cell nucleus, but that these territories can intermingle
with one another. Most functional genomic interactions
appear to occur in cis with neighbouring elements. How-
ever, many inter-chromosomal contacts have also been
documented, though the functional relevance of such
contacts (referred to as ‘chromosome kissing’) is still
unclear. Most contacts will inevitably arise because
chromosomes happen to lie next to each other or because
genes share common machineries such as those required
for transcription and splicing. Such neighbouring contacts
are likely to be important in chromosomal rearrangements
and some appear to have specific regulatory functions.
Ana Pombo and Miguel R. Branco use their expertise in
imaging techniques to further review in detail such inter-
chromosomal interactions. The first high-throughput
mapping of chromosome architecture in specific cell types
is currently in progress and such maps may help our
understanding of the mechanisms by which genome
architecture regulate gene expression. They discuss
how the (linear) genome may segregate into domains
that are more or less permissive for transcription within
the three-dimensional nucleus, and how the position of a
gene within the nucleus can enhance its activation or
silencing or even the efficiency by which its products are
processed or transported to the cytoplasm.
As a number of recent reports suggest that inter-chromo-
somal DNA interactions mediate the decision of which
allele to activate and which to silence, Wouter de Laat
and Frank Grosveld review functional inter-chromosomal
contacts in the context of mono-allelic gene expression.
They discuss these findings in relation to our knowledge
on gene competition, chromatin dynamics and nuclear
organization. They argue that both microscopy and bio-
chemical (4C) data strongly support the idea whereby
chromatin folds according to self-organizing principles
and that the nuclear positioning of a given locus is
probabilistic because it also depends on the properties
of neighbouring DNA segments. They further argue that
this stochastic concept of nuclear organization implies
that tissue-specific interactions between two selected
loci, present on different chromosomes, will be rare.
Finally, Ivan Rodriguez reviews the transcriptional con-
trol of the largest mammalian gene family, the odorant
receptor genes. The members of this family are located in
large gene clusters, located on different chromosomes and
are often interspersed by other genes and regulatory
Current Opinion in Genetics & Development 2007, 17:369–372
372 Differentiation and gene regulation
regions. Interestingly, olfactory sensory neurons express
only a single odorant receptor gene from a single parental
allele and represent a most extreme form of monoallelic
expression. How this is achieved is unknown, but recent
work points to multiple regulatory mechanisms, that may
also involve the interaction between loci located on
different chromosomes.
Conclusions
As these reviews illustrate, there is rapid progress in the
analysis of the behaviour of the genome. With increas-
ingly more ‘high-throughput’ whole genome techniques
and more sensitive proteomics, in particular mass spec-
trometry, the analysis of individual proteins can be unra-
velled more quickly and in more detail. In parallel,
improved imaging via new microscopic techniques
(particularly in the range of 5–200 nm, for review see
Cremer and Cremer [1]) and more sophisticated and
bright labels for visualisation are rapidly being developed.
Finally, the improvements on the informatics and math-
ematical side to process, interpret and model the vast
Current Opinion in Genetics & Development 2007, 17:369–372
amount of data quickly is expected to lead to even more
rapid progress in our understanding of the structure and
dynamics of the genome.
Acknowledgements
FG and DD are members of — and supported by — the EU FP6
programme ‘Cells into Organs’.
References
1. Cremer T, Cremer C: Rise, fall and resurrection of chromosome
territories: a historical perspective. Part II. Fall and
resurrection of chromosome territories during the
1950s to 1980s. Part III. Chromosome territories and the
functional nuclear architecture: experiments and models
from the 1990s to the present. Eur J Histochem 2006,
50(4):223-272.
2. Dekker J, Rippe K, Dekker M, Kleckner N: Capturing
chromosome conformation. Science 2002, 295(5558):1306-
1311.
3. Tolhuis B, Palstra RJ, Splinter E, Grosveld F, de Laat W: Looping
and interaction between hypersensitive sites in the active
beta-globin locus. Mol Cell 2002, 10(6):1453-1465.
4. Wilson EB: The chromosomes in relation to the determination
of sex in insects. Science 1905, 22(564):500-502.
www.sciencedirect.com