|
||||||||
|
IMMUNOLOGY A Mechanism Discovered for Antibody DeploymentApparatus May Contribute to Some Cancer Genes Now what? That's the question a mature B cell faces, lounging around a lymph node or spleen, when the right pathogen comes along. After its immunoglobulin (Ig) molecules recognize a specific antigen, the activated B cells and the Ig's they disperse need to know where to go and what to do.
In two papers, Katrin Chua, Ming Tian, Frederick Alt, Jayanta Chaudhuri (clockwise from left), and their colleagues show how mature, activated B cells take the first steps to alter their genome to make antibody molecules that know where to go and how to fight a specified pathogen. (Photo by Steve Gilbert) The deployment of antibodies in the immune response is directed by an ad hoc genetic shuffling and mutation--and scientists have just discovered how this happens. The findings are reported in complementary papers in the April 9 online Nature (April 17 in print) and the April 7 online Nature Immunology (May 2003 in print) by researchers in the lab of Howard Hughes investigator Frederick Alt. The discoveries, bolstered by related papers from two other labs published at the same time, "represent a quantum leap in our understanding," write Howard Hughes investigators Sebastian Fugmann and David Schatz, immunobiologists at Yale University, in a perspective in the May 2003 Nature Immunology. "Each Ig can recognize only a few highly specific structures," Fugmann and Schatz write. "But even the rare Ig that does find a good match, say on the surface of an invading virus, is typically not yet specialized enough to protect the host effectively against the infection. Inside the B cell, the genes that encode the Ig must undergo two refining processes to allow the Ig to acquire its full protective potential. The recent outpouring of discoveries provides substantial insight into the mechanism." B Is for BreedingB cells emerge from their development like debutantes, uniquely prepped in the bone marrow with one of 50,000 or more random variations of Y-shaped antibodies decorating the surface with pathogen-specific prongs. But despite the apparent diversity, all Ig's belong to only one class until they meet their matching pathogen.
"Class switching doesn't change antibody specificity to the pathogen, but it changes the way the body uses the antibody," said Alt, the Charles A. Janeway professor of pediatrics and genetics at HMS, Children's Hospital, and the Center for Blood Research. "It tells the antibody, once it binds to the antigen, what downstream pathways it uses to eliminate the pathogen." In each activated B cell, where Alt and his colleagues are looking, antibody genes rearrange themselves in the nucleus, designing and refining targeted antibodies. Their rapid mutation rate would be shocking anywhere else in the genome.
The diagram illustrates the central mechanism, class switch recombination, which begins with immune signals that set up two transcription sites in the "switch" (S) regions of the larger "constant" region. RNA sticks to the C-rich template strand long enough for the crucial enzyme AID to swoop down to the vulnerable single-stranded R-loop and change Cs to U's. When the DNA is back in its double helix, misfitting U's trigger DNA repair machinery, setting off a chain of events that eventually breaks the double-stranded DNA in the two switch regions. The DNA strand reconnects the two switch regions and forms a gene ready to make a new class of Ig's, while the excised region floats away. Each Ig gene can be separated into two distinct regions. The "variable" region tells the antibodies what pathogen to see. During B lymphocyte development in the bone marrow, three groups of coding sequences, known as variable, diversity, and joining, or V(D)J, randomly recombine to make exons that encode the unique two-pronged antigen receptors for mature B cells. In an activated B lymphocyte, these variable region sequences can be mutated to refine the Ig antigen specificity. Alt and his colleagues have worked out many of the details of this somatic gene assembly over the past 20 years. Further down the gene is the long "constant" region that tells the antibodies where to go and how to fight the pathogen. Here, a dozen exons in humans (fewer in mice) are separated by long "switch" (or S) regions, composed largely of cytosine (C) bases on the DNA template strand and companion guanines (Gs) on the nontranscribed strand. Following the first switch region, the first exon codes for IgM, setting up the default antibody class. In an activated B cell, the constant region for class M is permanently excised from the genome. Class switch recombination occurs between the switch region upstream of the M class exon and the switch region upstream of the target constant exon. "We wanted to try to answer three major questions about class switch recombination," Alt said. "Why do you need the switch region with the funny G- and C-rich strands as targets for the recombination process? Why do you have to transcribe through the switch region to make it work? What does the enzyme activation-induced deaminase (AID) do in the whole process?" Needed Slack in the SystemDuring transcription of the Ig gene switch region, when the DNA is unwound and unzipped, RNA tends to stick to the numerous Cs in the switching region of the template strand. That leaves the nontranscribed G-rich strand dangling out of the RNA-DNA complex in a temporary structure, the R-loop. The loop is a necessary template for class switch recombination in Ig molecules, according to the Nature Immunology paper.In a series of studies in genetically manipulated mice, led by postdoctoral fellows Reiko Shinkura and Ming Tian, the researchers first inverted an endogenous S region so that the G-rich strand became the new transcription template. Class switch recombination dropped substantially, apparently because the RNA does not stick to the Gs, so the R-loop does not form. Then the researchers replaced the endogenous switch region with an artificial sequence that forms an R-loop during transcription. It supported a significant level of class switch recombination. When they inverted the artificial sequence so the G-rich strand was transcribed instead, again, class switch recombination dropped significantly. Three years ago, Japanese researcher Tasuku Honjo showed that mice needed a putative RNA editing enzyme called AID to switch classes. Without it, they were stuck with an IgM tail on their antibodies. (Without AID, mice also cannot mutate the variable region of the gene to make more specific pathogen sensors.) In addition to identifying the critical enzyme, Honjo unleashed controversy by proposing that AID worked on the RNA part of the structure. Soon, British researcher Michael Noiberger countered this model with evidence from bacterial studies that AID mutated DNA rather than RNA. That's where postdoctoral fellow Jayanta Chaudhuri came in. In biochemical studies reported in Nature, he and his colleagues in Alt's lab showed AID mutates single-stranded DNA, but apparently not RNA or double-stranded DNA. Using the artificial sequence, he showed that AID attacks the exposed G-rich loop to convert Cs to uracils (U's). When he inverted the artificial sequence, AID did not convert the Cs, showing that the RNA-DNA loop transcription structure is a necessary template for AID's crucial role. "Uracils cannot form proper Watson-Crick base pairs with Gs and are thus recognized and processed by the base excision report machinery," Fugmann and Schatz write. "The subsequent steps leading to the generation of DNA double-strand breaks and finally to the completion of class switch recombination remain enigmatic." Two other papers in the May 2003 Nature Immunology from other labs confirm aspects of the Nature paper. Alt's lab is following up on those downstream mechanisms and also investigating how AID works to cause B cell lymphomas, myelomas, and other cancers, which can involve a translocation of a switch region into an oncogene. "All we know from current work is how the process gets initiated," Alt said. "Now we have a really good framework for figuring out what happens downstream." --Carol Cruzan Morton |
