Apr 142010
ResearchBlogging.orgAlthough we are most familiar with the circadian rhythm from its effects on our physiological state, the roots of the phenomenon lie in the molecular biology of individual cells. The circadian rhythm is the result of a transcriptional control system that regulates the levels of many different proteins in the cell with the passing of time. Not all of the proteins subject to this control have yet been catalogued, and as a result some surprising effects are still being discovered. A recent article in Proceedings of the National Academy of Sciences from the Sancar lab at UNC suggests that circadian control of a DNA repair factor may be a way to enhance the effectiveness of a chemotherapeutic agent. The article is open access, so I encourage you to open it up and read along.

Previously, the Sancar group has shown that the circadian rhythm affects DNA repair in brain cells. In that case, DNA had been damaged by UV irradiation, a lesion that had to be replaced by the excision repair mechanism. Because one of the critical factors for this kind of repair, the Xeroderma Pigmentosum A protein (XPA), undergoes a circadian oscillation, the efficiency of repair depends on the time of day at which the damage occurred. Kang et al. hypothesized that this circadian dependence could also be true for other forms of DNA damage that undergo excision repair.

This led them to cisplatin, a drug used as primary chemotherapy or part of a combinatorial regimen for several kinds of cancer. Cisplatin creates covalent bonds crosslinking DNA bases in an intra-strand or inter-strand manner. These covalent linkages make replication (and therefore mitosis) impossible, and elicit a DNA repair response. If the cell cannot clear the crosslink, it will either die by apoptosis, or if the apoptotic response is suppressed, fail to produce viable daughter cells. Either way, the growth of the tumor is suppressed. The excision-repair pathway is the only mechanism of clearing this kind of DNA damage, so Kang et al. thought that there might be a robust circadian dependence. In order to test this idea, they carried out experiments using extracts from mouse liver and testis.

Figure 1 shows the results of their experiment using liver extract. Panel C summarizes the key results (data shown in panels A and B) that XPA mRNA, XPA protein, and excision repair efficiency are correlated with the dark/light cycle the mice are experiencing, with the highest levels of protein and repair occurring in the late afternoon, and the lowest levels in the very early morning (around 5 AM). Panel E shows a comparison of excision repair between normal mice and ones that have been genetically modified to lack cryptochrome, a critical circadian clock protein. In these CryDKO mice the time of day has no effect on the efficiency of repair, and as panel F shows, they also do not have the daily fluctuation in XPA mRNA and protein levels. These results suggest that circadian control of XPA expression levels dictates repair efficiency in liver tissues — as panel D shows, addition of exogenous XPA protein can recover the excision repair activity. However, there was no detectable circadian dependence for excision repair activity in testis, as Figure 2 shows using similar experiments.

Circadian control of XPA activity is possible because the protein doesn’t last very long in the cell. Figure 3 shows experiments using two inhibitors: cycloheximide (CHX), which prevents protein synthesis, and MG132, which prevents protein degradation by the proteasome. The gel in panel A shows that in two different types of cells, CHX treatment caused XPA protein to disappear over a period of three hours, in contrast to the control protein actin, which was unaffected. Addition of MG132 caused XPA to accumulate with time, although actin levels were again constant. The fairly rapid degradation of XPA protein means that the overall quantity of that protein in a cell will be highly dependent on the concentration of its mRNA transcript. That is, you can only have high XPA levels if there’s a lot of mRNA so ribosomes can continue to produce it. Circadian control of transcription is therefore able to regulate protein concentration. The authors hypothesize that this tight circadian control may have developed to prevent deleterious effects of non-specific DNA repair activity during times when there is no chance of UV insult, although there is no specific evidence to support this conjecture presently.

The MG132 experiment indicates that XPA is degraded by ubiquitin ligation and subsequent destruction in the proteasome. Figures 4 and 5 show a variety of evidence indicating that this process is mediated by the ubiquitin ligase HERC2. The experiments in figure 4 establish that HERC2 binds to XPA (panels A and B) and colocalizes with it in the cell (panels C and D). The gels in Figure 5 show that when HERC2 levels in the cell are knocked down by RNA inhibition using siRNA specific for that protein, CHX treatment does not cause XPA to degrade. These facts indicate that HERC2 is the ubiquitin ligase for XPA. The direct effect of HERC2 activity on the repair of cisplatin DNA damage is shown by figure 6. Here, cells were treated with cisplatin and HERC2 siRNA. The left side of panel A shows the clearance of cisplatin adducts from A549 cells over time (the right side shows the total DNA). As you can see, addition of HERC2 siRNA allows for more rapid clearance of the adduct, with a particularly dramatic effect at low dosage.

Unfortunately, due to its extreme toxicity, treating cancer cells with CHX is not a viable strategy for chemotherapy. However, knowing that the level of XPA protein in some target cells varies in a predictable way with the time of day can help doctors optimize treatments for maximum effectiveness. In particular, for cancers originating in tissues that have a strong circadian rhythm and intact XPA, early-morning treatment with cisplatin may be more effective than treatment at other times of the day. More experiments are needed before this can be formally recommended — in particular, whole-animal studies and human trials will be necessary to definitively establish the effect. If these results hold up in whole organisms, however, the circadian effect on DNA repair may become a valuable tool for optimizing some chemotherapy regimens.

Kang, T., Lindsey-Boltz, L., Reardon, J., & Sancar, A. (2010). “Circadian control of XPA and excision repair of cisplatin-DNA damage by cryptochrome and HERC2 ubiquitin ligase”. Proceedings of the National Academy of Sciences, 107 (11), 4890-4895 DOI: 10.1073/pnas.0915085107

Full Disclosure: I have previously collaborated with Aziz and his group on a research project on eukaryotic cryptochromes.

Aug 232008
ResearchBlogging.orgThe ability to sense and respond to magnetic fields is a fundamental aspect of behavior in many animals. While migratory birds famously use the earth’s magnetic field to navigate during, magnetic field responses occur in all manner of animals, from eels to invertebrates. Even the lowly fruit fly, best known as a reminder that you really should have taken the garbage out two days ago, can react to magnetism. While various explanations have been put forward in different species, magnetosensitivity remains fairly mysterious. In this week’s Nature, researchers from the University of Massachusetts Medical School show that the blue-light photoreceptor cryptochrome plays an essential role in allowing fruit flies to detect magnetic fields.

Cryptochrome (or Cry) inherited the ability to receive blue light along with its photolyase domain, which is homologous to a prokaryotic, light-dependent DNA repair protein. Cry proteins, which are present in all animals, do not perform any DNA repair work, but instead play a role in regulating the circadian rhythm. While it is not clear in all cases whether Cry’s ability to absorb blue light is biologically significant in clock regulation, it is known that fruit flies (Drosophila melanogaster) use Cry to synchronize their circadian clocks. Previous experiments had suggested that the ability of fruit flies to detect magnetic fields was somehow related to photoreception, and that short wavelengths (like those sensed by Cry) had different effects from longer ones.

Gegear et al. devised a relatively simple experiment to test the importance of Cry in Drosophila magnetosensing. They placed a T-junction in a box, with a magnetic coil on one side and a non-magnetic coil on the other. They released flies into the junction, with (trained) or without (naive) performing an earlier run where the magnetic field was associated with a sucrose reward. They shined a light into the box and used filters to investigate the role of specific wavelengths.

They discovered that several strains of Drosophila could be trained to go to the magnetic field, although the degree of preference and the nature of the naive response differed substantially between strains. Gegear et al. chose the strain that showed the greatest response in full-spectrum light (and displayed a tendency to avoid the magnetic field in the naive state) to perform the filter experiment. Cutting off all wavelengths of light shorter than 500 nm abolished both the naive and trained responses to the magnetic field in these flies, as did filtering out all wavelengths shorter than 420 nm. If only wavelengths shorter than 400 nm were cut off, some of the trained and naive response returned. Simply dimming the light was not enough to replicate the effect of filtering. These experiments indicate that magnetic sensitivity in these flies requires light in the blue to ultraviolet range.

In order to prove that cryptochrome specifically is necessary for this magnetic sensitivity, Gegear et al. took advantage of our tremendous knowledge of fly genetic manipulation to create mutant flies that did not have a functional Cry gene. No matter what wavelengths of light were used in the T-junction experiment, these flies did not respond to the magnetic field. Crossing these Cry-null mutants with normal flies restored magnetosensitivity. The authors also performed experiments to show that the circadian rhythm was not itself essential to magnetic response in the flies.

Because this is a genetic experiment, it cannot address the question of whether Cry is both the blue-light photoreceptor and the magnetosensor. Going just on what we have in this paper, it is also possible that Cry acts upstream of another magnetosensor protein or is part of its downstream signaling pathway. However, in light of research that shows the flavin photoreception in other cryptochromes induces the formation of magnetically-sensitive radicals, some of which I discussed last year, it certainly seems possible that Drosophila cryptochrome does the whole job itself. As I mentioned in the case of the previous article, though, there is not yet any understanding of a mechanism by which information about magnetic field could be transduced from Cry radicals into the nervous system.

Dorosophila Cry differs from other plant and animal Cry proteins in significant ways, so it’s unclear whether these results have any relevance for other organisms. However, the finding that Cry is essential to Drosophila magnetosensitivity suggests at least the possibility of parallel systems in migratory birds and other species that use magnetic fields.

Robert J. Gegear, Amy Casselman, Scott Waddell, Steven M. Reppert (2008). Cryptochrome mediates light-dependent magnetosensitivity in Drosophila Nature, 454 (7207), 1014-1018 DOI: 10.1038/nature07183