This preview shows page 1. Sign up to view the full content.
Unformatted text preview: Midterm will be like the sample questions on the website. Covers up until this Thursday. The midterm is on Tuesday. Q&A Review Monday Y 6:30PM to 8:30PM CS 76 TA-led RNA editing Only uridienes are inserted, sometimes deleted. They are inserted and deleted at certain sites that actually makes ORFs and can be translated. They're very GC rich areas since it lacks Ts. We had sequienced the maxicircles. The other did the maxicircles for T. Brucei. There were 3 genes. ND7, CO3, and ATP subunit A6. They were totally missing in T. brucei. The tothers were there. Except we had gaps here and there and those gaps were very G rich. We wondered where they were. It was in the same place but hidden. It was homologous to other N7s. It was hard enough to explain a few Us but it was hundreds of Us at many sites. People got really interested in this. We didn't understand where the information was. We didnt'k now why it was only U. We coined it as Pan-editing. There's a gradation, simple editing, and then 5' editing, then this pan editing. These are the missing genes. You see all the genes are there in the same positions. The green is encoded and the red is added by editing. There are two domains of editing. Domain = region of editing. Some are totally pan-edited. The fact that GC rich was pre-editing there had to be some editing. We saw all these G rich regions that were totally in this region. When we got the editing we got the ND1, CO1, ND9, some are still unknown. We had a lot more edited genes. And then there was this quandry. We had this influx of sequnce information to these transcripts. Where did it come from? Some people even started to doubt the genetic dogma. We never see protein going back. But so they started to look at short sequences that could base pair with the U. None of them were perfect matches. so they looked at the top hits at the mismatches. If we allowed G to pair not only with C but with U, there was perfect matching. GU which is known as non-Watson Crick bp is weak but a bonafide base pair. Here is the edited RNA. The Us added are shown as little Us. We found the short sequences and they were all from the maxicircles and they were in between genes and so we made oligonts for these genes and hybridized a northern and found these short RNAs and sequenced them and they actually came from the genes. So we coined "guide RNAs" because they were guiding the editing. He is a U (right).... Other lab shows that editing goes from 3'end to 5'end. So let's start at 3' end. They're all specified a and then a g with an asterisk because the a or g guides the Us. These are only in the guide RNAs. They correct the mistakes in the message. Here's another one, There's also some GU base pairs that are included and then you see a complete match, a complete duplex. What's the difference between uppercase and lowercase? I just put lowercase to indicate the guiding nucleotides. The uppercase is actually in the genome. Where does the guiding one come from? I'll tell you. Here's an example, Cytochrome b. ....l There's a 3' end of the duplex and the 5' end. It's called the anchor duplex and that's where the guide is told to start editing. The guide RNA is here and it finds a complementary sequence. Some of the anchors are ... There are small proteins that can catalyze annealing. I left a whole where there should be u added. The guiding A should specify U. What happens now is the guid RNA hybridizes and somehow it repairs by adding Us at each editing site. In sucession finally giving a perfect sequence that gives a perfect duplex. There's no crystal structure but there is ... a GU bp is weak. Let's go back to the linearized maxicircle. I've indicated the guidRNAs that we've discovered. There's one upstream of 12S, etc. The ones that are boxed. These are not all the possible gRNAs that we need, but these are some of them. when we knew there were small RNAs shown by the Northern blot. 50-55nt long. So we ended labled 5'. P32 at 5' end. Ran a gel. To our surprise we saw a whole range of band that was one nt apart that formed a bell shaped. That means they are added 1 nt at a time to the 3' end. When we sequenced these, we found out they were all Us and they were not in the gene. The gRNAs also have Us at the end of their tale and there could be quite a few of them. We think one possible function of the Us is to stabilize the structure. The pre-edited region is GC rich. The tale cancome together and give a bulge. So the tail stabilizes. As we add the Us, we get this perfect gRNA hybride. We think it could tether. We had found this at least 5 years before and didn't know it. Maybe they encoded tRNAs. In our mito, there is a complete tRNA set. But in the tryp mito genome, there are no tRNA genes. Let's label and hybridize to blot of total RNA and we got a whole series of bands, one nt apart. When we first saw the bands, we had published a paper that encode small RNAs unsually. COIII, we had a guid for the 3' end but we didn't have a gRNA for the first 8 sites. It was incomplete. Dotted section. Luckily Sturm found a gRNA for sites 1-8. That told us that minicircles encode gRNAs. We had studied minicircle replication but we didn't know what they did. Some are in the maxicircles but most are in the minicircles. They have conserved region in CSB-3 with is an origin of replication. This is in conserved in every minicircles. Within the variable region is a gRNA. And the distance between the variable and conserved regions is the same. There's a bent region but we dont' konw what it's for. Crithidia has 2 conserved regions. CSB3. a gRNA and a bend at a certain distance. T. brucie has 1 CSB3 and 3 gRNAs all in the same direction. At the ends of the gRNAs, there are ____ inverted repeats that would form a hairpin. Between them there's a guide. But T. brucei has 3 gRNAs per minicircle and Leishmania 1 gRNA. Cruzi has 4 gRNA genes. It's as iff there was a repeating structure that was duplicated. Cruzi can duplicate 4gRNA per circle. Remember the evolutionary tree? all the _____atids are all pathogens and mainly studied because they're in human diseases. There's probably a single ancestor to the entire family of trypanoplastids and that's where catenated minicircles arose. Then there were the bedonids but branched off, they were precursors of trypanochromatids and that's this T. borreli from a deep sea fish. We found 2 species. 180K bp that encoded gRNAs in tandem. Instead of minicircles, we had a large one with a bunch of gRNAs. We constructed that minicircles came from this. Then we thought something like maxicircle that were catenated. As soon as we found minicircles and gRNAs, we had the mechanism. it's very simple. Pre edited region PER. Here's the U tail. Here are the extra guide nts. The gRNA find the message that needs to be edited and hybridizes and forms an anchor duplex. Let's say there's a nuclease that cuts a mismatch. ssNuclease. Now we have a 3'end liberated and we have 2 fragments of message. 3' and a 5'. Let there be an enzyme that adds U. It's known as a 3' terminal Uriadyl transferase. TUTase. If we add 3Us they'll bond with aga. Then let there be a ligase that allows them to ligate together.. so we've extended it. So we have two extra As here. We go through the same cycle. Breaks, adds, ligates, breaks etc. U tail hanging out. The gRNAs specifying the RNA editing. Then we saidk, how does deletion work? Same way with an additional thing. Let's say we have an enzyme that recognizes a bulge. It can recognized the structure of the message. it cuts there, 3'-5' exonuclease who trims only Us. We have 3 Us hanging off with no bp, so let there be an enzyme that deletes 3 Us because the information is lacking in the gRNA. We call this enzyme cascade, some call it splice. This guy says, the U is put in by the gRNA. The mech is just like RNA splicing. It were insert the U by transferase. This became the flavor of the week. RNA editing was very ancient. Whereas cut/splice had to had enzymes. Then there was refuting papers. Beautiful theory hit hard facts. So the old theory was correct. This just shows, from our first paper. Celavage, addition, ligation, that's one cycle. Each cycle is one U added. It explains why editing goes from 3' to 5'. that's the first editing site. Editing goes 3' to 5' end within an editing block. Block: done by one gRNA. Domain is by multiple gRNA. Refuted: Misediting, a lot of mistakes made and they are corrected and when the true one is done, it works. They were due to the wrong gRNA forming a false anchor. It would bind, go along and stop because the next one wouldn't work. within a domain, multiple guides, this is ATPase 6. The 1st gRNA comes in, forms an anchor, the secone gRNA can hybridize and it edits up to here, the next one comes in and forms an anchor with THAT edited RNA so we get 6 gRNAs. The overall polarity is 3'5' because there are new anchors made. I have 2 dots where there are 3Us. Mostly on the 5' end. What has to happen? When it finishes editing, it has to have single strand. DNA breathes. It's weakly basepaired. There's less energy holding this together. The minute breathes or unwinds here, it forms an anchor to the unedited sequence There is the open reading frame that is created until a methioneine formed by a AUG and then it stops and finishes. Now it's homologous. Here's the difference between guides. The anchors can very from small to large. the gRNAs can be ... it varies and quite high. One gRNA can add a whole bunch of Us by multiple There's some deletions too. Where does RNA editing occur? In the cell. Is it localized in the mito to one specific region? We addressed that with immunofluorescence. We put the cells on a slide and did indirect immunofluorescence. DAPI, filters, we could see the kinetoplasts and the nucleus. Kinetoplast is the green and the small blue dot. Mito tracker is red. It's the exact same distribution as the green. There's a concentration of editing in the ____ but it could be _____ too. How did editing evolve? They have a crypto gene and an editing gene? Why?? Maybe in the early ancestors, that gave rise to the whole rise of kinetoplastids. Maybe there were mistakes. We know DNA pol makes mistakes at Ts. Maybe they were missing Ts in genes and the cells would die. They need these enzymes to survive. But preexisting enzymes like TUTase etc It used to add the information. The cell would survive and this would be the selective pressure. Editing rose to due to mistakes in ... Added where the errors are at the RNA level. What would fix this the editing phenomenon? What's different about the information transfer? GU and GC, we have a wobble. When you allow GU and GC, you fix the process. The gRNAs are no longer antisense so they can't go back in. Nope, nvm. I said that wrong. The idea is the complementary of the reverse strand because it can't be due to the wobble, but if you make a fully edited molecule and there's a RT that can go back because it has the Us. We looked all these species and looked at A6. We have 6 gRNAs all overlaping. Crithidia has 4, Herpetomonas has 4, cruzi has more, brucei has more. When wel looked at this, we said, there's a restoration of sequence 3'5'. If you look at the sequence, it looks like it's going one block at a time. Crithidia, we've lsto 2. In Cuicis, we've lost 5. In the other wones, there's no editin.g So you can lose editing. What is used for initiation of translation? There's a U inserted between A and G, if you don't have that it won't go. MAYBE they editing to regulate mitochondria biosynthetic editing. Like in blood serum there's no mitochondria DNA. But in Tsetse there's fully functional mito. How can this occur in evolution? We're assuming pan-editing was an ancient character? We say there are minicircles. In the top right, it has to be minicircles. We're looking loss within trypanosoma, there's a gRNA for block 1. Then 2 has gRNA for block 2, etc. That's the model. What would happen if gRNAIII were lsot? Minicircles.. Remember they're randomized throught the entire circle and replicate and pull otu and get antipodal. We get a randomization of mini circles. How many sequence classes are there? We' have a tarentolae that hasn't seen a gecko for 18 years. An old lab ___. Some classes were 1000, some 10000, some as few as 10. how does it force this process? We did computer simulations and made a model and let there be random distribution. What would happen to different sequence classes. For 1000s of cycles, the frequency of each class fluctuates. What if we did essential vs. non-essential. Non-essential ones became the main ones. They could be lost and die. Perhaps due to random segregation there is random loss of min ...
View Full Document
- Fall '08