For much of my professional career as a Drosophila geneticist I've worked with polytene chromosomes, and it's always interesting to see papers with interesting tidbits of information about their structure and function. Polytene chromosomes are those rather strange structures formed from high levels of chromosome endoreduplication (see this blog article for a detailed description). Polytene chromosomes are widespread in flies, and in Drosophila are mostly studied in the larval salivary glands where they are easy to work with: Calvin Bridges used salivary gland polytene chromosomes to construct his polytene chromosome map. In this paper, Tom Hartl and colleagues show Condensin complexes (which have a function in chromosome condensation and anaphase chromosome segregation; and in vitro can induce and trap DNA supercoiling) can cause polytene chromosome disassembly and antagonise transvection. Their data link processes of chromosome condensation and DNA supecoiling with higher order interphase nuclear structure that impacts on gene expression.
Unlike salivary gland polytene chromosomes, those of ovarian nurse cells break down during the development of the nurse cells, at about mid-oogenesis. In this paper, two mutant alleles of a predicted component of the condensin II complex, Cap-H2 are studied. In flies mutant for Cap-H2, the nurse cell polytene chromosomes don't disassemble.
In the figure (may require subscription) you can see the normal stage 10 nurse cell nucleus (A), in which the polytene chromosomes have disassembled. In contrast, in nurse cell nuclei of flies mutant for Cap-H2, polytene chromosomes persist (B-D). This suggests that condensin complexes (or at least Cap-H2) are required for the tight synapsis of polytene chromosomes to break down. So what effect does Cap-H2 have on the classic polytene chromosomes of the salivary gland? In contrast to those of nurse cells, salivary glad chromosomes persist. One might therefore expect that loss of Cap-H2 by mutation would have little effect, but that over-expression of Cap-H2 would bring about disassembly of salivary gland polytene chromosomes. This is indeed what was seen, and in fact can be clearly seen in two supplementary videos (Quicktime format). In the first movie, a three dimensional image of a salivary gland polytene nucleus is shown. The DNA (and therefore chromatin) is stained in red, while the green blob is an in situ hybridisation signal, demonstrating the tight synapsis of that locus. The second movie shows a salivary gland nucleus in which polytene synapsis has broekn down upon over-expression of Cap-H2. Here, no trace of polytene structure can be seen, while the in situ hybridisation signal resolves to numerous non-synapsed dots.
The functions of condensin complexes in assembling and maintaining DNA supercoiling is thought to contribute to overall chromosome structure, and in particular maintaining the domain structure important for long-range regulation of gene expression. This is where Cap-H2 mutants might reveal a role for Cap-H2 in transvection. In flies, it's thought that homologous chromosomes remain tightly paired through interphase. Transvection is a genetic phenomenon of cross-talk between such paired homologous chromosomes, and can typically be revealed by combinations of mutations within the regulatory and coding regions of genes. Here, two examples are examined - the Ultrabithorax (Ubx) and yellow (y) genes.
This is a really elegant part of the paper, so despite the fact that transvection is something I've always found difficult to explain, I'll try and do it here! UbxCbx-1 is a dominant mutation that causes a malformation of the wing-blade. This is because the mutation affects sections of the gene that control its expression. On the other hand, Ubx1 is a loss of function mutation within the Ubx gene. You might expect, therefore that a fly bearing both these mutations in the same copy of the Ubx gene might have normal wings. In actual fact, as can be seen in the figure below, this is not the case - compare the class D wing blade with the class A wild type wing blade - you can see the posterior margin of the blade is wrinkled and deformed.
This occurs because the dominant mutation UbxCbx-1 is able to control the expression of the wild type copy of the Ubx gene carried on the homologous chromosome, a feat it is able to achieve because the homologous chromosomes are closely paired during interphase. You can see that this is the case by comparing the class D panel of the figure with the class B panel, in which the pairing of the two chromosomes is disrupted by a chromosomal rearrangement - consequently, the wing blade morphology is closer to wild type.
So this is an example of transvection - cross-talk between homologous chromosomes. We've seen above that the condensin II complex component Cap-H2 is required for polytene chromosome dissolution in nurse cell development, and that inappropriate over expression in salivary gland cells caused the the polytene chromosomes to disassemble. Does loss of Cap-H2 affect the chromosome pairing in normal interphase, as judged by an impact on transvection? That this does indeed occur is shown by the increased strength of the UbxCbx-1 Ubx1 phenotype when the fly is also mutant for Cap-H2. This combination results in wing blades like that shown as class E in the figure - a severe malformation is seen. Further confimation that this is due to a transvection effect is provided by the class C wing blade, where a restoration of virtually wild type morphology is brought about by disruption of chromosome pairing by a chromosome rearrangement.
This evidence for a role of Condensin II complex components in interphase chromosome pairing and transvection, is backed up by similar experiments using y, mutants in which make the fly's body less pigmented than normal. But I'll leave that part of the paper for you to read!
The authors speculate that condensins mediate supercoiling activity not just when chromosomes are condensing for mitosis, but also in the interphase nuclei, where this plays an important role in the higher order structure.
T. A. Hartl, H. F. Smith, G. Bosco (2008). Chromosome Alignment and Transvection Are Antagonized by Condensin II Science, 322 (5906), 1384-1387 DOI: 10.1126/science.1164216