Researchers in the laboratory of Frederick Alt at the Program in Cellular and Molecular Medicine at Children's Hospital Boston and the Immune Disease Institute (PCMM/IDI) have determined that the RNA exosome, known to be a cellular 'quality control' mechanism for RNA transcription, makes crucial contributions to immunity.
Reporting in Cell (February 4, 2011), co-first authors Uttiya Basu, Fei-Long Meng, Celia Keim, and their colleagues at the PCMM/IDI, Columbia University College of Physicians and Surgeons, the Memorial Sloan-Kettering Cancer Center, and Rockefeller University have identified the RNA exosome as a long-theorized co-factor that gives activation-induced cytidine deaminase (AID) access to both strands of DNA, bringing about mutations and creating diversity among antibodies.
Before presenting an overview of the new findings from the Alt lab, let's review some relevant genetic and immunologic concepts; some will be familiar, and others may hearken back to science courses of long ago.
DNA's two-stranded double helix, which encodes our genome, is familiar to most of us. In replication, the entire molecule is 'unzipped' and copied. However, when a protein is to be made, a gene in a single section of the DNA is opened up and an enzyme called RNA polymerase uses one of the DNA strands as a template to "transcribe" a piece of protein-encoding RNA.
The Alt lab has a long track record of work on lymphocyte development and activation, including two B cell-based immunoglobulin (Ig) diversification processes: somatic hypermutation (SHM), in which antibody genes are mutated at up to a million times the normal genomic rate, vastly increasing immune specificity; and immunoglobulin heavy chain (IgH) class switch recombination (CSR), in which the Ig produced by a B cell is changed from one immune type to another.
Both SHM and CSR depend not only on transcription, but also on mutation. This is where the AID enzyme enters the picture. Sometimes referred to as a master regulator of antibody diversification, AID initiates both SHM and CSR by causing mutations in both the template and the non-template DNA strands of target genes.
Another important player to introduce is the RNA exosome, a tightly bound complex of 9 core protein subunits in the cellular nucleus that processes and degrades (with the assistance of two additional catalytic sub-units and co-factors) flawed or unnecessary RNA transcripts.
Dr. Alt points out the central question: "For both SHM and CSR, a huge body of research clearly confirms that AID must access and mutate both the template and the non-template strands of DNA. But our chromosomal DNA exists in a duplex form; whereas biochemically AID works only on single-strand DNA. How does AID get access to both duplex DNA strands in B cells?
To determine what might actually be giving AID access to the template DNA, the team transcribed duplex DNA in a test tube using a bacterial polymerase and extracts from human B-cell lymphomas; when they purified the AID-associated transcription complexes, they found subunits of the RNA exosome. When they omitted the polymerase, AID did not associate with the exosome and had no effect.
Then co-first authors Uttiya Basu and Fei-Long Meng immunoprecipitated different components of the exosome complex from cells. Regardless of which subunit was targeted, all the other subunits and AID were pulled down with it, not only confirming the tight coherence of the exosome's subunits-but also revealing the strength of the association between AID and the exosome.
If the RNA exosome really does guide the action of AID, as suggested by the biochemical experiments, the Alt team suspected that they should find it on the IgH switch regions of B cells activated for CSR, which are the AID targets in this process. In confirming this, they further discovered that AID itself has a role in attracting the RNA exosome.
In a series of biochemical assays detecting deamination of each DNA strand separately, the group determined that the RNA exosome stimulated AID to deaminate both transcribed DNA strands. They were able to conclude, "Thus the RNA exosome is a long-sought co-factor that can target AID activity to both template and non-template strands of transcribed SHM and CSR targets".
The group further hypothesized that the exosome catalytic subunits might give AID access by degrading RNA bound to the template strand and exposing it to AID. However, in testing this hypothesis, they found that the reconstituted 9 sub-unit core of the exosome, purified as individual components after expression in bacteria, stimulated AID activity on both strands even without the catalytic sub-units, indicating potentially novel RNA exosome functions. However, the group emphasized that these biochemical studies still leave open a possible role for the exosome catalytic subunits in cells.
Dr. Alt explains that though the new findings relied on in vivo knockdown and other modern genetic/genomic techniques, his team also had to resort to some 'old-school' lab work: "You can't knock out some essential genes, like those that encode RNA exosome components, or you'll kill the cell. Thus, you can't always do definitive in vivo experiments with them. So if you want to find out how such proteins might work, you've got to do classical biochemistry."
Though the lab made use of in vitro biochemical techniques and a bacterial polymerase, which can't perfectly reproduce what human RNA Polymerase (Pol) II does in living cells, the work does point to possibilities for the specifics of the exosome/AID interaction. Because the scientific literature offered no clear-cut model of this molecular mechanism, the Alt lab produced a working model based on the known characteristics of AID, the RNA exosome, and CSR. Dr. Meng kindly produced a simplified version of the published graphic to accompany the explanation below.
However, before we examine that model, Dr. Alt introduces us to another player already familiar to his lab: "We did some biochemistry some years ago and showed that if AID is phosphorylated, it can bind a cellular replication protein called RPA, which in turn binds single-stranded DNA. If you put the AID/RPA complex on DNA substrates and transcribe them in vitro, you can also target AID activity to the non-template strand." (Recall that SHM and CSR act upon the both the template and non-template DNA strands.)
The proposed model shows the double-stranded DNA (dsDNA) as a blue ladder: two solid lines connected by rungs. Single strands of DNA are black lines. As the Pol II (shown here as a white fishing boat) moves down the DNA generating a transcript (an orange line with a large orange oval, the ribonucleoprotein or RNP), it stalls and backtracks. This not only releases the transcript but also attracts both the RNA exosome (represented as a large fish, swallowing the newly produced RNA) and AID (orange balls, with small black 'U's showing deamination products), which in turn associates with RPA (large green ovals) along the single-strand DNA. RPA may stabilize the ssDNA newly separated by the exosome to give AID access.
Dr. Alt reminds us that this is a working model, designed to account for only what is known thus far; much remains to be learned. He places the new findings in context: "The biology of AID is a hot field, because it is important not only for generating antibodies, but it is also showing up as sort of a 'bad actor' that generates mutations that lead to many different types of tumors, and perhaps an important player in the whole stem cell biology/genomic reprogramming area. We'll have to understand how AID gets targeted to proper targets to understand how it gets targeted to the wrong places. So knowing how it really works could affect a lot of fields."
Dr. Meng readily suggests what's next for this line of research. "We are trying to map how AID and the RNA exosome move together and exactly how AID recruits exosomes to the switch region. Why AID functions specifically on Ig loci is also very interesting, with a lot of related questions about how various factors focus on a particular gene".
"We also do not fully understand how AID is recruited and bound, and downstream, how AID-associated factors like RPA may recruit DNA repair pathways. I think our future work may help connect those pathways. We have already created a mutant mouse model in which AID constitutively binds to RPA. Maybe we'll get some exciting results."
Basu U, Meng FL, Keim C, Grinstein V, Pefanis E, Eccleston J, Zhang T, Myers D, Wasserman CR, Wesemann DR, Januszyk K, Gregory RI, Deng H, Lima CD, Alt FW. The RNA Exosome Targets the AID Cytidine Deaminase to Both Strands of Transcribed Duplex DNA Substrates. Cell. 2011 Feb 4;144(3):353-63. Epub 2011 Jan 20.
