#circularrnas

There are also a few more biological functions of circular RNAs that have been uncovered so far. Transcription regulation is especially common amongst ciRNAs and EIciRNAs with introns retained (basically circular RNAs with introns can regulate transcription). They regulate transcription by interacting with the U1 snRNP and RNA Polymerase II.

Alternative splicing can result from the biogenesis (creation) of circular RNAs, so another role of circular RNAs is to help facilitate alternative splicing through their own creation. This implies that the biogenesis of circular RNAs from a pre-mRNA strand is actually competing against and dysregulating the normal canonical pre-mRNA splicing in order to facilitate alternative splicing.

Another role of circular RNAs is to regulate parental gene translation. They do this if the circular RNA contains the AUG start codon for translation, so the remaining exons from the RNA that the circular RNA was made out of won’t have a start codon, and thus won’t get translated. This indirectly reduces the translation of parental genes.

Some circular RNAs also help to regulate progression through the cell cycle. One circular RNA known to do this is circ-Foxo3 (forkhead box O3). The circular RNA Foxo3 binds to p21 and CDK2 to regulate progression through the cell cycle. P21 is a protein that inhibits CDK2, and CDK2 is a cyclin dependent kinase (enzyme), which is supposed to be active in only the G1 and S phase and needs to be shut off for mitosis to proceed.

Circular RNAs can also be a protein sponge. CircMBL binds to the muscleblind protein (MBNL) and sponges out any extra MBNL proteins. The MBNL protein was shown to be ‘implicated’ in myotonic dystrophy which causes muscle loss and weakness (I’m still not too sure if too much MBNL is good or bad). Some other examples of RNA-Binding Proteins that can bind to circular RNAs are argonaute (AGO1, AGO2) which is needed for gene silencing and micro RNA processing, EIF4A3 (eukaryotic initiation factor) which is a translation initiation factor #4A3 (i.e. starts translation), FMR1 (fragile X mental retardation) which is a polysome-associated protein (a polysome is a cluster of ribosomes held together by the mRNA that they’re translating) that helps neuronal development and synaptic plasticity by regulating mRNA splicing and mRNA stability. There is also RNA demethylase ALKBH5, and other proteins which can be sponged by circular RNAs. You can imagine that there are many proteins capable of being sponged by compatible circular RNAs, so these protein sponges can end up affecting all the functions that these proteins were supposed to carry out such as gene silencing, miRNA processing, translation initiation, neuronal development, etc.

Another role of circular RNAs is to help facilitate m6A-driven translation. m6A is N6-methyladenosine. If a circular RNA contains an RRACH sequence with m6A, then that circular RNA can drive the initiation of translation. R is a purine (adenine or guanine), A is m6A (N6-methyladenosine), C is cytosine, and H is a non-guanine base (adenine, cytosine, or uracil). Basically, this just means that we can translate the circular RNA into protein products if it has an m6A inside a RRACH sequence. How this works is, MTTL3 and MTTL14 form a heterodimer (the two join together. MTTL is MeThylTransferase-Like protein), and they methylate an adenosine. FTO stands for alpha-ketoglutarate-dependent dioxygenase, and FTO can come and de-methylate a methylated circular RNA. So, after the MTTL complex methylates the circular RNA at an adenosine, the YTHDF3 protein can come and bind to the m6A, and then this recruits the EIF4G2 (eukaryotic translation initiation factor), which also recruits the 43S preinitiation complex (made of a bunch of eIF proteins) and then translation can occur. Another way to translate circular RNAs is actually through an Internal Ribosomal Entry Site (IRES). A circular RNA could have an IRES motif sticking out of it, which allows RNA Binding Proteins to latch onto the IRES motif and go along the circular RNA. Some microRNAs and silencing RNAs (siRNAs) are also involved in this mechanism.

Another function of circular RNAs is to act as a microRNA sponge. In order to be a miRNA sponge, the circular RNA must have a microRNA Response Element (MRE) which lets the circular RNA bind micro RNAs and thus reduce binding of microRNAs to their intended targets. Normally, miRNAs tend to bind to argonaute proteins. Even though it was demonstrated that some specific circular RNAs have the MRE which allows them to bind miRNAs, in actuality the majority of circular RNAs DO NOT have the ability to act as miRNA sponges.

In total, the field of endogenous circular RNAs is a new and very fast-moving field of research. If there are any problems with the RNA splicing mechanism in the body, then it will be reflected by the generation of circular RNAs. Thus, a last possible function of circular RNAs is to act as a disease biomarker. Circular RNAs are pretty stable inside a cell due to being able to resist exonucleases. Plus, circular RNAs are normally expressed in certain cell and tissue types. So not only can circular RNAs be used as a diagnostic tool for specific cancers or diseases, they might also be able to become a therapeutic tool. It’s possible to use silencing RNAs (siRNA) to target specific circular RNAs for downregulation.

For example, the circular CCDC66 and CiRS-7 RNAs are found in colorectal cancers (CRC). It’s also possible to find circular RNAs in bodily fluids such as saliva and in plasma, which makes them easy to collect from patients. Circular RNAs are also stable in exosomes (basically a bunch of vesicles that are inside one larger vesicle that gets fused into the membrane so all the smaller vesicles inside are released outside). These exosomes contain tumor-specific circular RNA and are released by tumor cells in order to protect the circular RNAs from RNases (enzymes that degrade RNA) found in extracellular fluid (fluid outside the cells, like your blood or whatever). When these circular RNAs inside exosomes enter the blood, we can draw them out and measure their presence in blood for cancer detection. This can allow us to identify patients with early stage colon cancer through only a blood test.

The production of circular RNAs results from the mRNA splicing that occurs in eukaryotic cells. The eukaryotic spliceosome is technically what cuts out the circular intronic RNA (ciRNA). Usually ciRNAs result from a lariat intron (intron that has a branch point with the A). After the splicing, the lariat intron sometimes evades the normal process of debranching and degradation via an exonuclease (i.e. the lariat is usually cut up and destroyed). Instead, the 3’ tail hanging out of the lariat loop structure ends up getting trimmed and results in a stable circular intronic RNA (ciRNA).

Another way that circular RNAs can be produced is through backsplicing. For example, if we have exons 1 to 5 with introns in between each of them, we could end up having the end of exon 1 joining directly with exon 5 during the splice, so the spliced out portion still contains exons 2, 3, and 4. Then we can have splicing occur on this portion on only the ends of exons 2 and 4, and then the exons 2 and 4 end up joining together to form a circular RNA with exon 3 still inside the circular RNA sequence.

For a more detailed example, let’s consider this sequence of exons. So, pretend we have two exons with an intron in between them.

(Head of Exon1) AG----GT (Tail of Exon1) (Intron) (Head of Exon2) AG----GT (Tail of Exon2)

In linear splicing, the tail of exon1 joins with the head of exon2. Whereas in circularization, the tail of exon2 joins with the head of exon1. This is possible because when splicing, all we need is a GT sequence which acts as the donor (attacker) and an AG sequence which acts as the acceptor. In circularization, the tail end of exon2 loops around and attacks the head end of exon1.

Biogenesis of Different Types of CircRNAs

There are four types of circular RNAs. Exonic circRNAs (ecircRNA), intronic circRNAs (ciRNA), exonic-intronic circRNAs (EIciRNA), and transfer RNAs (tricRNA). Although backsplicing doesn’t occur very frequently, there are ways to increase the frequency of backsplicing and thus increase the amount of circular RNA that is generated. For example, there could be reverse complementary sequences on the precursor RNA strand which bind together and form a loop which brings downstream introns and exons close to the upstream RNA. In addition to reverse complementary sequences, RNA-Binding Proteins (RBPs) can also hold RNA together in a loop-around to promote backsplicing.

Before, the viroids and virusoids we talked about were all exogenous circular RNAs, meaning that they came from outside the cell. There are also endogenous circular RNAs which come from inside the cell.

First let’s go through a short overview of the history of circular RNA research. In 1976, the first circular RNA was discovered. This was a viroid. In 1979, circular RNAs were first observed in eukaryotic cells (cells of multi-cellular organisms, i.e. not bacteria). In 1986, circular RNA was first identified in humans. This was called the Hepatitis Delta Virus. In 1988, we found that circular RNA was exceptionally stable. In 1991, the first endogenous circular RNA was identified in humans. This meant that even without some kind of infection, it’s still possible for the human body to produce circular RNA on its own. In 1995, we created new circular RNA and found that it could be translated in vitro. This means that it might be possible to manufacture circular RNA and deliver it into cells and then encode it to produce the proteins that we need to produce for whatever treatment. We also found that inverted repeat sequences were necessary on the RNA in order to circularize it. In 1998, we further confirmed that engineered circular RNA could be translated in vivo. In vivo means in a living organism while in vitro tests are done on cell cultures that we grow but are not inside a living organism. This in vivo test further raised the viability of developing circular RNAs as drug treatments.

Virusoids are satellite circular RNAs. In virology, the term ‘satellite’ is used to describe infectious particles which require a helper virus in order to replicate after infecting the host. Thus, you usually cannot get infected by a virusoid alone if you aren’t infected by the right helper virus, and even if the virusoid manages to get into your cell, it won’t have the necessary equipment to replicate and thus spread in your body. The ‘equipment’ that virusoids need from their helper virus is the RNA polymerase. Virusoids are very similar to viroids but virusoids require the RNA polymerase from a helper virus instead of how viroids just use the RNA polymerase that the host cell uses.

Virusoids are usually 220 to 338 nucleotides long, so they are still very short circular RNAs. The virusoids contain ribozymes, which is like what the avsunviroidae use for self-splicing their concatemers during replication. The helper virus that virusoids need is a sobemovirus (a genus of virus which infects plants). One other difference between virusoids and viroids is that virusoids are encapsidated (packaged) by the coat protein of their helper virus, while viroids are just naked RNA. However, a similarity is that the majority of virusoids also have no coding capacity due to their small size.

Some examples of virusoids are the Lucerne transient streak virus satellite RNA (virusoid with LTSV as the helper virus), and the Subterranean clover mottle virus satellite RNA. Usually the trend is to prefix the name of the helper virus and then add ‘satellite RNA’ to the end to describe the virusoid. In Professor AbouHaidar’s lab, they discovered the smallest and only known exogenous coding and replicating circular RNA, the Rice Yellow Mottle Virus satellite RNA. This is a virusoid that is 220 nucleotides in length which also has coding capacity in the genome.

Ribozymes

Virusoids self-splice during replication with their own ribozymes. These are sections of their RNA that splice themselves out of the RNA chain and then go around and keep on cutting out more copies of the virusoid. Since replication of single stranded RNA needs to go from plus-sense to minus-sense to plus-sense again, there are both plus and minus-sense ribozymes. The particular ribozymes used for virusoids is called a hammerhead ribozyme, because it has a central box-like area with no base pairing and three stems of base-pairing RNA coming out of it. The stems are positioned so that one stem is the ‘handle’ of the hammer and the other two stems are roughly opposite to each other so they represent the face and claw of the hammer. These hammerhead ribozymes need magnesium (2+) ions in order to self-splice.

2 Magnesium (2+) ions are required for the ribozyme to perform its functions. No proteins or ATP is required for the ribozyme to do its thing, so even if you shut down the cellular translation or energy-producing functions, these virusoids can still replicate as long as there are magnesium ions. The ribozyme performs self-splicing by having one of the magnesium ions attack the hydroxyl group of the ribose, and then the other magnesium ion receives the charge from the phosphate backbone, which ends up severing the phosphate backbone link between the next RNA. The reverse of that reaction can also be catalyzed by the hammerhead ribozyme, and the result of that is self-ligation (re-joining the ends of the cut-off strands into a circle).

Apart from the hammerhead ribozymes which are present in many plant virusoids and viroids, there are also other types of ribozymes. Group I Introns are a type of self-splicing intron (introns are sequences of RNA that are removed during splicing. Only exons make it into the final RNA) which is present in pre-ribosomal RNA. Usually, introns wait for a spliceosome to come and cut them out, but these self-splicing introns can cut themselves out of the sequence. Then there are Group II Introns, which are also self-splicing introns that have been found in bacteria and the organelle genes of eukaryotic cells. This ties into how we suspect that eukaryotes originally might have evolved from swallowing prokaryotic bacteria which then became the organelles.

Another ribozyme is called RNase P. Ribonuclease P catalyzes site-specific hydrolysis of precursor tRNA which is essential for the formation of tRNA. What this means is that the RNase P cuts up immature tRNA (when we make new tRNA) at certain places, and that helps it to become fully functional mature tRNA.

Then we have the Hepatitis Delta Virus Ribozyme. This is the ribozyme that the HDV virus uses, and it is a noncoding RNA that is thought to be the only catalytic RNA known to be required for viability of a human pathogen. I.e. the HDV virus cannot function without its HDV Ribozyme. Remember that the virusoids and viroids which also use their own ribozymes all infect plants, so that is why the HDV Ribozyme can carry the title of the only one in humans.

Many of the characteristics of viroids was investigated in trials done on potato spindle tuber viroids (PSTV) since it was the first discovered viroid. These experiments were done by T.O. Diener in late 1969.

When he tested PSTV with endonucleases (cuts inside the strand), he found that incubating PSTV with RNases destroyed the infectivity of the viroid while incubation with DNases did nothing. Thus, from the endonuclease tests we could conclude that the viroid genome consists of RNA.

For exonucleases, he tested both Tobacco Mosaic Virus RNA (TMV RNA) molecules and PSTV viroid particles. The enzymes used were snake venom phosphodiesterase (SVPD) and phosphatase. Each RNA underwent three types of trials: incubation with only snake venom phosphodiesterase, with only phosphatase, and with both enzymes.

For the Tobacco Mosaic Virus RNA trials, incubation with only phosphatase did nothing, but the trials which included snake venom phosphodiesterase ended up canceling out the infectivity of the tobacco mosaic virus.

Meanwhile, for the Potato Spindle Tuber Viroid trials, all three types of trials did not curb the infectivity of the viroid. The inefficacy of exonucleases on PSTV signified that it was a circular genome, and the results with endonucleases told us that PSTV was an RNA. Put together, we could conclude that PSTV is a circular RNA molecule.

Viroids can be classified into two main groups: the Pospiviroidae and the Avsunviroidae.

Pospiviroidae

The pospiviroidae include the previously mentioned potato spindle tuber viroid (PSTVd). Pospiviroidae replicate in the nucleus of the plant host cells. Since they replicate in the nucleus, it makes sense that they would take advantage of the cellular machinery present there. Pospiviroidae use cellular ribonucleases to cleave its RNA and cellular RNA ligases for joining its RNA. This way, RNA processing for Pospiviroidae is done completely by host cell machinery.

Using our example of the potato spindle tuber viroid (PSTVd) that we discussed before, we can split the pospiviroidae genome into various regions. As discussed before, the PSTVd viral genome is one single circular RNA but may look like a double stranded RNA due to the extensive amount of base-pairing happening when the strand loops back on itself. This pseudo-linear genome has been given these labels for its regions from left (where nucleotide 1 and 359, the two ‘ends’ join together) to right (middle of the actual RNA sequence, where it loops back on itself): TL, P, C, V, TR

The TL and TR stand for the terminal left and terminal right regions. These represent the left and right ‘ends’ of the RNA loop respectively. The P stands for the Pathogenic region (plays a role in infectivity), C for the Conserved region (doesn’t change much), and V for the Variable region (region where mutations are more likely to happen). One interesting thing to note is that the nucleotide numbers for these regions is not actually continuous because the RNA loops around. For example, the TL contains approximately nucleotides 1 to 40 and 320 to 359. The pathogenic regions has around nucleotides 40 to 80 and 280 to 320, etc. One exception is the terminal right region where the viroid RNA loops, so the terminal right region contains nucleotides 150 to 210 (the top and bottom half of the region is continuous).

Avsunviroidae

The name avsunviroidae comes from one of its representative viroids, the avocado sun blotch viroid. Another viroid in this avsunviroidae family is the peach latent mosaic viroid. In contrast to how the pospiviroidae replicate in the nucleus, these avsunviroidae replicate in the plant chloroplasts instead. Due to the lack of nuclear host machinery such as the ribonucleases and RNA ligases, they instead have a self-cleaving RNA. This is called a ribozyme, which means an RNA molecule that can function similar to a protein enzyme. This ribozyme carries out the all the necessary cleaving of viroid RNA in lieu of the ribonucleases.

Potato Spindle Tuber Disease (PSTVd)

The potato spindle tuber viroid (PSTVd) was the first discovered replicating circular RNA. The term replicating is important, because many circular RNAs are produced normally by both plant and animal cells during splicing. However, the viroids that we are talking about are different from those kinds of circular RNAs because viroids replicate.

At first this potato spindle tuber disease was reported in potatoes in the 1920s. Potato tubers that were infected with PSTVd looked like a spindle instead of being round like they’re supposed to be. Infected potatoes are shaped more like cucumbers but are still potato-like in color and texture. This symptom was reported to slowly spread through potato fields, which signified that it was probably caused by some kind of infectious agent.

Although it is named the potato spindle tuber disease, this is mainly because when it was first recognized as a disease, the host happened to be potatoes. Thus is ended up being named that way. In reality, it can infect a variety of other crops such as tomatoes and avocados. Laboratory experiments using PSTVd has also shown it to infect many other crops.

Transmission of PSTVd occurs either from direct contact in the field or by mechanical transmission in the lab (i.e. you take a scalpel, scrape an infected potato, then use the same scalpel to cut another tomato). No arthropods are involved (no insects, arachnids, etc.). It also does not undergo soil-borne transmission (does not spread through the soil). However, PSTVd can be transmitted through pollen and seed.

Cadang-Cadang Disease (CCCVd)