Peer pressure often evokes unexpected behavior, and protein partners might do the same. The Methyl-CpG binding protein 2 (MeCP2) has long been considered a transcriptional repressor, given its penchant for binding histone deacetylases and methylated stretches of DNA. But in today’s Science, researchers led by Huda Zoghbi at Baylor College of Medicine, Houston, Texas, report that MeCP2 turns on more than five times as many genes as it turns off. This revelation has important implications for the study of Rett syndrome, autism, and other neurological conditions that have been linked to MeCP2 mutations. “The bottom line is that Rett syndrome, caused by the loss of this protein, is really a disease of loss of activation rather than activation. That is what is fundamentally important to focus on,” said Zoghbi in an interview with ARF. Exactly what makes MeCP2 an activator is not clear, but the work suggests that other protein partners may facilitate the process.
MeCP2 fell under the spotlight when scientists discovered its role in Rett syndrome. This is a neurodevelopmental condition that manifests itself early in childhood, and mostly in girls (the MeCP2 gene lies on the X chromosome, and in boys, loss-of-function mutations in the single good copy of the gene can prove fatal early in development). Though it was once assumed that de-repression of genes was the key to Rett syndrome and other MeCP2-linked disorders, it was later shown that having too much MeCP2 is just as bad (Van Esch et al., 2005). Why that is has been unclear. One possibility is that an abundance of MeCP2 is de-facto loss-of-function because the excess protein sequesters binding proteins that are needed for transcriptional suppression. To test that idea, Zoghbi and colleagues looked at gene expression profiles in the hypothalamus of MeCP2-negative and MeCP2-overexpressing mice. If the profiles turned out to be the same in both, then the loss-of-function theory would hold water.
“The big surprise was that the expression profiles were in totally the opposite direction,” said Zoghbi. The second surprise was that most of the genes are activated, not repressed, by MeCP2. When first author Maria Chahrour compared gene expression profiles of four wild-type mice with those from four animals overexpressing MeCP2 (MeCP2 Tg) she found that the majority of genes tested (2,184 out of 2,582, or 85 percent) were upregulated in the latter. In four MeCP2 nulls, those very same genes are downregulated. The researchers validated the profiling analysis by using quantitative amplification to measure mRNA levels of 66 of the most significantly altered genes.
This exact correlation between activation and repression suggests that many of the affected genes are directly regulated by MeCP2 rather than some secondary effect of MeCP2 toxicity. To test this, the researchers used chromatin immunoprecipitation experiments to see if MeCP2 binds to the promoter regions of six specific genes. All four genes that are activated (somatostatin, opioid receptor κ, guanidinoacetate methyltransferase, and G protein-regulated inducer of neuritic outgrowth) and two genes that are repressed (myocyte enhancer factor 2C and ataxin 2 binding protein 1) by MeCP2 bound the protein, though Zoghbi said that it is hard to know exactly how many of the 2,000 plus genes are directly affected. “If one was to pick 20 genes, then I don’t expect all of them to bind [MeCP2],” she said. “But irrespective of how many of those activated genes are direct targets, we think this work is really important,” she added. Michael Greenberg and colleagues at Children’s Hospital, Boston, agree. In an accompanying Science Perspective, they note “the ability of Chahrour et al. to correlate changes in the expression of particular genes with the level of MeCP2 expression convincingly links these transcriptional changes to the presence or absence of MeCP2.”
How does a protein that appears to have all the hallmarks of a repressor activate transcription? Zoghbi said that she does not see MeCP2 as a transcription factor in the traditional sense of binding to specific DNA sequences, but rather a modulator that guides activators to critical sites on DNA. In fact, Chahrour and colleagues found that CREB1, a major neuronal transcriptional activator, co-purified with MeCP2. Sequential chromatin immunoprecipitation experiments also showed that both proteins bind to the promoter of the somatostatin gene, one of the genes activated by MeCP2 according to the profiling analysis. And when the researchers used the somatostatin promoter to drive expression of luciferase in neuronal cells, they found that expressing both CREB1 and MeCP2 led to more luciferase activity than when either was expressed alone. The findings suggest that the dual binding of the proteins to the somatostatin promoter is functionally significant.
Zoghbi and colleagues proposed that MeCP2 is a bit of a chameleon, suppressing gene transcription when in the company of histone deacetylases or other DNA silencing proteins such as Sin3A, and activating transcription when partnering with CREB1 or other transcription factors. She noted that DNA context may also enter the equation. “I think MeCP2 is still a repressor, but I wouldn’t be surprised if it repressed a lot of promoters at repetitive sequences and transposable elements, rather than neuronal or cell-specific genes,” she suggested.
Of the more than 2,000 genes altered when MeCP2 is perturbed, finding those that directly bind to the protein will be the next big research push. What is interesting, Zoghbi suggested, is that the protein’s widespread effect on transcription is mirrored in its widespread effect on development. Indeed, she has come to view Rett syndrome like a conglomerate of many neurological diseases. For example, cognitive impairment, rigidity, chorea are all present in Rett patients, and have parallels to Alzheimer, Parkinson, and Huntington diseases. “If you pick any symptom of any neurologic disease and look at all the patients that have a mutation in the MeCP2 gene, they have it. I think that is striking,” said Zoghbi, who originally trained as a neurologist before adding genetics and molecular biology. Because of this broad distribution of symptoms, Zoghbi suggested that subtle mutations that affect MeCP2 levels could put people at risk for Alzheimer’s or another neurologic disease later in life. In fact, Zoghbi and others recently reported that mutations that halve MeCP2 protein levels in mice lead to subtle neurological symptoms (see Samaco et al., 2008 and Kerr et al., 2008). “No one has yet looked at human mutations that might affect the level of protein, i.e., mutations that affect the promoter, post-transcriptional regulation, and perhaps post-translation modification. But thinking about the big picture, I wouldn’t be surprised, given how important this protein is for synapses, for example, if having a variation that might leave someone with 70 percent of this protein rather than 100 percent, might make them susceptible later on in life to late-onset, sporadic neurodegenerative disease,” Zoghbi said.
In other Rett syndrome news this week, researchers from the University of Massachusetts, Amherst, further report on just why some of the mutations in MeCP2 are so damaging. Christopher Woodcock and colleagues looked at the effect of missense mutations on the structure of MeCP2. First author Rajarshi Ghosh and colleagues found that three frequent mutations (R133C, F155S, and T158M) destabilize the methylated DNA binding domain (MBD) of MeCP2 both in solution and when bound to DNA. Furthermore, two mutations (R106W and F155S) abolish a subtle but significant increase in α-helical structure that accompanies DNA binding. The work suggests that a major effect of these Rett mutations is to prevent the protein from adopting its native structure or to bind properly to DNA targets.—Tom Fagan
- Van Esch H, Bauters M, Ignatius J, Jansen M, Raynaud M, Hollanders K, Lugtenberg D, Bienvenu T, Jensen LR, Gecz J, Moraine C, Marynen P, Fryns JP, Froyen G. Duplication of the MECP2 region is a frequent cause of severe mental retardation and progressive neurological symptoms in males. Am J Hum Genet. 2005 Sep;77(3):442-53. PubMed.
- Samaco RC, Fryer JD, Ren J, Fyffe S, Chao HT, Sun Y, Greer JJ, Zoghbi HY, Neul JL. A partial loss of function allele of methyl-CpG-binding protein 2 predicts a human neurodevelopmental syndrome. Hum Mol Genet. 2008 Jun 15;17(12):1718-27. PubMed.
- Kerr B, Alvarez-Saavedra M, Sáez MA, Saona A, Young JI. Defective body-weight regulation, motor control and abnormal social interactions in Mecp2 hypomorphic mice. Hum Mol Genet. 2008 Jun 15;17(12):1707-17. PubMed.
No Available Further Reading
- Ghosh RP, Horowitz-Scherer RA, Nikitina T, Gierasch LM, Woodcock CL. Rett syndrome-causing mutations in human MeCP2 result in diverse structural changes that impact folding and DNA interactions. J Biol Chem. 2008 Jul 18;283(29):20523-34. PubMed.
- Chahrour M, Jung SY, Shaw C, Zhou X, Wong ST, Qin J, Zoghbi HY. MeCP2, a key contributor to neurological disease, activates and represses transcription. Science. 2008 May 30;320(5880):1224-9. PubMed.
- Cohen S, Zhou Z, Greenberg ME. Medicine. Activating a repressor. Science. 2008 May 30;320(5880):1172-3. PubMed.