How a C. elegans Histone Methyltransferase Targets Regions to Silence
As an organism grows and develops, its cells are rapidly dividing and committing to different fates. Some, for example, will become immune cells, while others will become skin cells. What decides the different fates of these cells is how their genes are expressed. For example, to become a skin cell, a particular cell will need to express specific genes that will help it develop the features of a skin cell while also silencing genes that are not necessary. Genes are often silenced by methylation of the ninth lysine on histone H3 (H3K9me), which leads to the formation of heterochromatin, compacted and therefore, silenced DNA. This methylation is done by enzymes called histone methyltransferases (HMTs).
In C. elegans, an important model system for studying development, the histone methyltransferase MET-2 is responsible for adding either one or two methyl groups to H3K9 – called mono- or di-methylation, respectively. An open question remains, however, in how MET-2 is targeted to regions of the DNA that need to be silenced. New work by Delaney and colleagues in the Journal of Cell Biology demonstrates that MET-2 requires the protein LIN-65 to properly localize to heterochromatin and thus, to maintain proper gene silencing as cells commit to different fates.
To see where MET-2 normally localizes in the cell, the authors tagged MET-2 with a fluorescent mCherry tag and found that it localizes to the nucleus throughout embryonic and larval development. They saw that it clusters into foci within the nucleus. Next, they wanted to ask if these foci of MET-2 form on heterochromatin. Using gwIs4, a GFP-tagged marker for heterochromatin, the authors found that its signal overlapped with the MET-2 mCherry signal. Furthermore, both signals localized to the periphery of the nucleus, where heterochromatin is typically found.
What could be targeting MET-2 to heterochromatin? To answer this question, the authors looked to known proteins that read H3K9me marks on histones. These proteins, often called reader proteins, recognize different modifications on histones and contribute to their repression or activation . When the authors knocked down expression of known reader proteins, they saw no loss of MET-2 foci, indicating that these reader proteins are not important for the formation of MET-2 foci. In the case where the expression of reader protein CEC-4 was lost, they did see that while MET-2 foci remained present, not all of the foci localized to the periphery of the nucleus, indicating that MET-2 foci need CEC-4 for proper localization.
Because none of the known reader proteins were involved in MET-2 heterochromatin targeting, the authors used immunoprecipitation of MET-2 paired with mass spectrometry to identify proteins that tightly bind to MET-2 and therefore, might help target it. They found clear enrichment of the two proteins LIN-65 and ARLE-14. By immunoprecipitating either protein, they found clear enrichment for the other protein and MET-2, indicating that LIN-65 and ARLE-14 simultaneously bind MET-2. They also found that LIN-65 and ARLE-14 both localized in nuclear foci, which mostly co-localized the MET-2 foci.
Because both LIN-65 and ARLE-14 interact with MET-2, the authors next wanted to ask if these three proteins depended on each other. They found that if expression of LIN-65 was lost, MET-2 and ARLE-14 foci completely disappeared, but a weak fluorescent signal for each protein remained in the nucleus and cytoplasm, which suggests that LIN-65 is important for foci formation. Loss of ARLE-14, on the other hand, had no effect on MET-2 or LIN-65 foci. Loss of MET-2 resulted in loss of both LIN-65 ad ARLE-14 fluorescent signal, indicating that their stability depends on MET-2.
Clearly, the interaction between LIN-65 and MET-2 is important for MET-2 regulation, but is LIN-65 required for the histone methyltransferase function of MET-2? By assessing the level of H3K9me2 in both met-2 and lin-65 mutant embryos, the authors showed that there was an equal decrease in H3K9me2 in both mutant lines as compared to wild type. Moreover, both lin-65 and met-2 mutants had decreased fertility compared to wild type embryos, suggesting that MET-2 activity depends on the ability of LIN-65 to form foci in the nucleus.
The authors next asked how LIN-65 and MET-2 both affect gene silencing. They saw that when LIN-65 expression was knocked down by RNAi, there was a loss of repression of the heterochromatic reporter gwIs4. In fact, using RT-qPCR they found that loss of LIN-65 relieved expression of MET-2 target genes, while non-MET-2 targets were not affected. To get a better sense of how loss of LIN-65 or MET-2 effects transcriptional control across the genome, they performed RNAseq of lin-65 mutant, met-2 mutant, or the double mutant lin-65;met-2 embryos. They found that overall, the aberrantly expressed genes in lin-65 or met-2 mutants correlated significantly. They also observed that in met-2 mutants, there were de-repressed genes that were not found as de-repressed in lin-65 mutants, suggesting that LIN-65 represses gene expression almost completely through its interaction with MET-2. They authors suggest that LIN-65 enhances the association of MET-2 with heterochromatin and thus its activity there.
In addition to the authors’ findings that LIN-65 stabilizes and localizes MET-2 at heterochromatin, prior work has shown that LIN-65 is required for C. elegans’ response to high temperature stress . To ask if the role of LIN-65 in stress response is linked to a possible role for MET-2 in stress response, the authors subjected embryos to 37°C, a stressful temperature. They found that at 37°C both MET-2 and LIN-65 foci dispersed, but when shifted back to the non-stressful temperature of 20°C, both foci recovered. To ask if this dispersal of foci had an effect on the animal, they saw that loss of either MET-2 or LIN-65 resulted in a decreased number of progeny.
In their final experiment, the authors ask how MET-2, LIN-65, and the other C. elegans histone methyltransferase, SET-25 are involved with how heterochromatin is associated with the inner nuclear membrane (INM), located at the periphery of the nucleus. Prior work had shown that if either SET-25 or MET-2 expression were lost, there was no effect on heterochromatin localization to the INM, but if both were lost, heterochromatin no longer localized to the nuclear periphery . Using ChIP-seq for the protein LEM-2 which associates with the INM, they found that if SET-25 was lost, heterochromatin remained enriched there, but if MET-2 was lost, heterochromatin lost its association with the INM. The authors confirmed this result through DNA-FISH and found that the enrichment of heterochromatin in the INM was lost in met-2 and lin-65 mutants, but not in wild type or set-25 mutants. These results show that even though there is an additional histone methyltransferase present, H3K9me2 by MET-2 is required for proper heterochromatin anchoring at the nuclear periphery.
Taken together, this work identifies LIN-65 as a new regulator of MET-2 function in C. elegans. The authors show that LIN-65 interacts with MET-2 to form foci at heterochromatic regions in the nuclear periphery. By working together, these proteins ensure that the necessary genes are silenced during development, and in doing so leads us to a deeper understanding of how exactly our cells determine their fate.
Original Article: Delaney CE, Methot SP, Guidi M, Katic I, Gasser SM, Padeken J (2019). Heterochromatic foci and transcriptional repression by an unstructured MET-2/SETDB1 co-factor LIN-65. J Cell Biol, pii: jcb.201811038. doi: 10.1083/jcb.201811038.
 Musselman CA, Lalonde ME, Côté J, Kutateladze TG (2012). Perceiving the epigenetic landscape through histone readers. Nat Struct Mol Biol, 19(12):1218-27. doi: 10.1038/nsmb.2436.
 Tian Y, Garcia G, Bian Q, Steffen KK, Joe L, Wolff S, Meyer BJ, Dillin A (2016). Mitochondrial Stress Induces Chromatin Reorganization to Promote Longevity and UPR(mt). Cell, 165(5):1197-1208. doi: 10.1016/j.cell.2016.04.011.
 Towbin BD, González-Aguilera C, Sack R, Gaidatzis D, Kalck V, Meister P, Askjaer P, Gasser SM (2012). Step-wise methylation of histone H3K9 positions heterochromatin at the nuclear periphery. Cell, 150(5):934-47. doi: 10.1016/j.cell.2012.06.051.