Gene regulation

Kissing chromosomes

A three-dimensional examination of gene regulation suggests that portions from different chromosomes ‘communicate’ with each other, and bring related genes together in the nucleus to coordinate their expression.

The cell has evolved many strategies to orchestrate gene activation or repression. Spilianakis et al.1 (page 637 of this issue) reveal a novel mechanism of gene regulation, throwing light on how cells organize their genome to respond efficiently to stimuli. They show that genes on different chromosomes that are destined to be expressed within a common cell lineage are brought together in the nucleus. Such inter-chromosomal communication has been suspected for some time, but this is the first evidence that it actually takes place.

Our understanding of gene regulation has moved from an initial notion of a one-dimensional array of regulatory elements next to each other on the same thread of DNA as the gene (Fig. 1a) to an appreciation that genes are associated with groups of proteins, forming multimolecular complexes that are arranged in structures generically known as chromatin2. The subsequent discovery that distant, contiguous sequences can have a profound effect on gene expression introduced a second dimension onto the scene, with ‘looping’ and ‘scanning’ (probably mediated by the attached proteins) invoked to explain these long-range interactions (Fig. 1b)3.

Figure 1: The three dimensions of gene regulation.

a, Linear view of gene regulation. The promoter (P) near the start of a gene provides the minimal information needed for gene expression. The function of the promoter is supplemented by enhancers or silencers (E), farther away, where regulatory proteins bind to activate or repress transcription of the gene (arrow). b, The ‘looping-scanning’ model of gene regulation. The locus control region (L) regulates several genes. Proteins binding here scan through large portions of DNA, looping the intervening region out, until they find the relevant gene. c, Gene regulation in 3D. Spilianakis et al.1 find that genes from different chromosomes (A and B) are in close proximity until a developmental signal stimulates the cells, when the genes split apart. One moves to a region that represses gene expression (heterochromatin, red) and the other relocates to an area with many active genes (euchromatin, green). d, Genes from different chromosomes might come into contact when the chromatin containing them loops out from their chromosome ‘territory’.

But to explain how genes that are far removed from each other in the genome, and even on different chromosomes, can be coordinated to be expressed together, or to preclude the expression of one another, required a leap into a third dimension. The spatial location of a gene within a cell nucleus can determine whether it is expressed or not: genes residing in areas of chromatin that contain repressive factors (heterochromatin) are silent; conversely, genes in nuclear regions full of activating proteins (euchromatin) are usually switched on (Fig. 1c)4,5. How do genes find their appropriate location in the nucleus of a cell, and how are genes that must be expressed herded into active neighbourhoods?

To address these questions, Spilianakis and colleagues1 used immune cells called T cells. As they mature, T cells organize themselves into subsets that are assigned specific duties. Thus, T helper (TH) cells produce factors that help other cells of the immune system to function optimally. After antigen stimulation, naive (undecided) TH cells develop into either TH1 cells, which produce one set of effector molecules (for example interferon (IFN)-γ), or TH2 cells, which produce a different set (for example interleukin (IL)-4 and IL-5)6. The authors explored the organization of two genomic regions within the TH subsets: the gene encoding IFN-γ (called Ifng), which is mainly active in TH1 cells, and a multi-gene complex including the genes encoding IL-4 and IL-5 (Il4 and Il5), which is mainly active in TH2 cells.

Viewed through two-dimensional analyses, Ifng seems to be regulated by elements found near it on chromosome 10, whereas expression of Il4 and Il5 on chromosome 11 seems to be regulated by a ‘locus control region’ (LCR) on the same chromosome, which directs the entire TH2 gene complex. Spilianakis et al. asked whether these two genetic regions are in close proximity or interact in the nuclei of naive TH, TH1 or TH2 cells.

Biochemical and imaging experiments showed that the two regions (on two different chromosomes) are in close proximity in the nucleus of naive TH cells, but after stimulations that induce a TH1 or a TH2 state they seem to move away from each other (Fig. 1c). The authors' interpretation of this is that in the naive TH cell, the two gene complexes are close together in a region of the nucleus that is poised for gene expression. Upon receiving a specific stimulus, the gene to be activated (for example Ifng after a TH1 stimulus) is allowed to begin expression, whereas its counterpart that is to remain silent (in this case the TH2 genes) is moved, presumably to a more repressed region of the nucleus (Fig. 1c).

Spilianakis et al. provide some evidence that following stimulation, the inter-chromosomal interactions in the naive cell are replaced by intra-chromosomal ones between regulatory elements within the same gene region. Moreover, the inter-chromosomal association does seem to regulate to some extent the expression of the genes involved, as deletion of a regulatory element (HSS7) from the TH2 LCR on chromosome 11 disturbs the inter-chromosomal associations, and this is coupled with delayed expression kinetics of Ifng from chromosome 10.

These remarkable findings will puzzle us for some time to come. Are inter- and intra-chromosomal associations a general phenomenon occurring in all types of cell7 or not8? What are the mechanisms that bring two functionally related genes on two different chromosomes together? Between cell divisions, chromosomes expand to occupy ‘chromosomal territories’ in the nucleus9,10. To allow genes from different chromosomes to come into close proximity, therefore, perhaps chromatin strands might extend from the main body of a chromosome's territory to an area of the nucleus where transcription is possible (Fig. 1d)7,11.

Do both copies of the Ifng gene find their functional counterparts of the TH2 gene complex at some point in the cell's development programme? The results presented by Spilianakis et al. are snapshots and indicate that at any particular time the majority of the cells show only one copy of Ifng in close proximity to one TH2 complex. The remaining Ifng and TH2 complex are found separately. Does this mean that there is fluidity in this interaction, allowing genes to come together and then drift apart again after some time? Or does it imply that these genes are expressed from only one of their copies? To start addressing these questions, and others that similar systems may bring up, we would have to develop technologies that allow the detection of inter-chromosomal interactions not as a snapshot, but as a movie. Is it time to go 4D?


  1. 1

    Spilianakis, C. G., Lalioti, M. D., Town, T., Lee, G. R. & Flavell, R. A. Nature 435, 637–645 (2005).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Felsenfeld, G. & Groudine, M. Nature 421, 448–453 (2003).

    ADS  Article  Google Scholar 

  3. 3

    De Laat, W. & Grosveld, F. Chromosome Res. 11, 447–459 (2003).

    CAS  Article  Google Scholar 

  4. 4

    Brown, K. E. et al. Cell 3, 207–217 (1999).

    CAS  Google Scholar 

  5. 5

    Kioussis, D. & Festenstein, R. Curr. Opin. Genet. Dev. 7, 614–617 (1997).

    CAS  Article  Google Scholar 

  6. 6

    O'Garra, A. Immunity 8, 275–283 (1998).

    CAS  Article  Google Scholar 

  7. 7

    Osborne, C. S. et al. Nature Genet. 36, 1065–1071 (2004).

    CAS  Article  Google Scholar 

  8. 8

    Kim, S. H. et al. Cytogenet. Genome Res. 105, 292–301 (2004).

    CAS  Article  Google Scholar 

  9. 9

    Parada, L. A., Sotiriou, S. & Misteli, T. Exp. Cell Res. 296, 64–70 (2004).

    CAS  Article  Google Scholar 

  10. 10

    Cremer, T. & Cremer, C. Nature Rev. Genet 2, 292–301 (2001).

    CAS  Article  Google Scholar 

  11. 11

    Kosak, S. T. & Groudine, M. Genes Dev. 18, 1372–1384 (2004).

    Article  Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kioussis, D. Kissing chromosomes. Nature 435, 579–580 (2005).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.