This preview shows page 1. Sign up to view the full content.
Unformatted text preview: Structure and Properties of antibodies Dr. Morrison OH Tues 2PM Dr. Zack OH Mon 2PM Discussion section is mandatory, but we allow you to attend a different section if we check in with that section's TA and tell them our name and why we're there. MAke sure both TA's know. Lecture 1 was an overview. The last lecture will be an overview by Dr. Morrison. She'll be giving lectures on B cells. Dr. Zack will give lectures on T cells. Lecture 2: There's two parts to immunity, adaptive and innate. We'll focus on the adaptive immune response. It's devided in cell mediated and humoral. Humoral comes from the old ages where "humors" were circulating in your body. Humoral talks about circulating factors in your body. People started to do experiments with these factors. We won't be responsible for the people. Koch injected guinea pigs with V. cholerae and V. melchnikovvi. When the guinea pigs were exposed again to the same organism, they were protected but not differet though related ones. Now we know this is due to antibodies. If bacteria is injected in blood, we get substances that have different properties. If we inject and then take out the blood again, we find the bacteria is clumped. Sometimes you get opsonization, you get coated substances from serum and the phagocytic cells can better engulf. If you add to a soluble protein from that bacteria, you get a precipitate. TA's will go over the mechanisms. Important concept is that all of these properties is due to antibodies we have in the serum. Antibodies interchangeable with immunoglobulin. Antigen can be bound. All immunogens are antigens, but not all antigens are immunogens. Immunogens can elicit an immune response. The development of new techniques lead to a break through in understanding. They used electrophoresis. If you put proteins in an electrophoretic field, they will move according to MW or charge. They wanted to study what antibodies were. They immunized a rabbit and studied it and also studied the rabbit after adding albumin. They found it was an immunoglobulin that had this characteristic vibrating at gamma. They would get gamma globuliin to protect against polio for examplel. There was also ultracenttrifugation. They would observe, the larger the protein material, the faster it sediments. They purified these antibodies and analyzed in the ultracentrifuge. They came at a rate of 7S which is about 150k MW. They further analyzed and looked at the ratio of the precipitated antigen and antibodies. For every antibody, there were two antigens present. It has a valence of 2. To get some insight into the antibody structure. They took the enzyme, papain, and digested the antibodies and ran them on an ion exchange column and found two peaks. The Fab bound antigen. Fc could form crystals. For every Fc molecule, there were two Fab produced. Now they took pepsin and digested the antibody and found that he got a molecule Fab'2 would bind AND precipitate the antigen. Fab only bound it. If he treated this with something that broke a sulfide bond, then the Fab'2 would release two Fab. Another guy didn't cleave with pepsin and only broke the disulfide bonds. He separated them on a size exclusion column. There were two peaks, a heavy and a light chain. They began to put together structures of what they thought it looked like. Also, they like to inject antigens to make antiserum so they made antiserum to these moieties. If they have an immunogen that react with H chain then it would react with Fab and Fc. If it was for L chain, it would react only with Fab. But anti-Fab would work with H AND L. They figured it was like a heterodimer. They know the whole molecular weight is 150kMW. Look at diagram to see where the enzymes cut at the antibodies. The structure was consistent with the data. This is the fundamental building block of an antibody. This is how they actually look. The heterogeneity of antibodies is noticed. There's a nice sharp feak for albumin but the antibodies are broad: alpha, beta, gamma. It formed a problem for protein chemists. It was realized that when multiple myeloma was a malignancy of plasma cells which are antibody producint factors. Myeloma is a variant growth of a single clone. Therefore it's making a single species of Ig. It's alll homogenous to give a nice sharp peak. The graph on left shows different clones which makes it heterogenous. This is called polyclonal. Often in myeloma, these people make extra light chain. they are small and are filtered out through the kidney. These light chains are now monoclonal but also homogenous. So, when this was done, the protein analysis was very crude. How are antibodies capable of abinding many different antgens? It's because you can produce antibodies for almost anything. Even synthetic organic molecules. When we look at Ab, we find two different kinds of light chains, kappa and lambda. If you make a anti-serum for kappa, it will recognize all the kappa light chains but not lambda. Same with anti-lambda. What the protein chemists did was take advantage of these proteins and digested them with proteases. Then analyzed those products with 2-D peptide map. They looked at those different maps and compared them. If they looked compared to kappa, about half of those peptides are the same but the other half are different. If you compared to lambda, about half were shared adn half were the same between the two light chains. But if you compare kappa to lambda, nothing was the same. The conclusion was that light chains have two regions: one part that was variable and one that was constant. The next breakthrough was aa sequencing. They began to sequence these proteins. If you compare the sequence of a whole bunch of light chains. you find that the variability was at the N-terminus and the constant side was at the C- terminus. People began to sequence the heavy chains. It was very variable. There were several different opsions for the heavy chain but the light chain had only kappa or lambda. How do we determine the specificity of the variable region? If you look at these variable regions, he lined up all these sequences, some areas were conserved and some areas were hypervariable. He made a veriability chart. If all were equal, frequency would equal 400. If it was only one aa, it would be 1. He made all these plots and found that there were 3 regions within the H chain and 3 regions in the L chain where most of the variability was high. Also, there were the complementarity determining regsions. This is where it was important for binding to the antigens. The prediction is that the CDRs would contact the antigent. In addition, they also did crystallography of some of these proteins. If you looked at the crystal strucute of the light chain, they correspond to the aa forming the variable and constant domain. They realized that the CDRs were actually present on exposed loops at the N-terminus of these chains. The hypervariable regions were exposed. the less variable are structural ones to determine the fold of the protein. They all have the same conserved folding regions. It's the framework to hold the complementarity regions. People began to make crystal structures of the Lchains. They have two domains, the variable and the constant. Heavy chains have four or five domains (ie IgG have 5?) All antibodies are glycoproteins which means they have a carbohydrate moiety. Moiety means substance. The structure was made on mutant ab that didn't have a hinge. But recent studies of ab is that there is a region called the hinge. It allows the Fab to move to have different distances relative to the other Fab. it can now simultaneously bind to antigens. Dinitrophenyl (DNF) on two ends with a peptide spacer to immunoprecipitate Abs on electromicroscopy. They can precipitate with different orientations for their Fab. The barbells in the diagram top right are the DNFs. They can make trimers or tetramers. You'd have to have different orientations. That shows that there is something that allows them to flexibly move to recognize different antigens to allow fit. What's sticking out is the Fc. Fab is bound to make the shape. They digested with pepsin, it would cut off the Fc and leaving the Fab. We now see trimers and tetramers lacking the Fc. There are several different structures for the constant region of the H chain. They are divided in 5 classes. IgM, IgG, etc. IgG actually has 4 different kinds in circulation. Those are called subclasses. If we look at IgA, we find there are 2 different kinds. The number of subclasses and classes = isotype. Mu isogtypis IgM or IgG and the 4 subclasses are called gamma isotype. light chains have two isotypes. The different isotypes are encoded by a separate genes. Light chains genes aren't on the same but for H chains they are. IgM is made of 5 basic building blocks or 6 parts. If there are 5 copies, there is another molecule called J chain that is made by the same cell that makes the chains. If it's 6, then it doesn't need the J chain. They usually look like a starfish . We see a conformational change to look like a spider later. IgM is usually the first response. The initial immune response is of low affinity. IgM has all these different binding sites (10-12). It has high avidity, in terms of multiple binding site, can compensate. Affinity = is the actual number of association constant. It describes the nature of the interaction. Avidity means the summation of all the binding together. Low affinity it comes on and off. But because of high avidity, the sum of all the low affinity. It is also called functional affinity. All the binding sites are identical on a single IgM. Binding through multiple sites. IgG is the most abundant. It also has a half life of 23 days. it can cross the placenta. It provides protection ot the newborn. It binds Fcgamma receptors. It will also activate complement. IgA is also a polomeric polymer. It can be a monomer of a dimer held together with a J chain (joining) or can be a secretory IgA where a 4th chain is associated with the J chain and the dimer. J chain and dimers are all from one cell but the the 4th chain is from a secretory cell. IT is the most abundant immunoglobulin. It comes from submucosal cells. It is a critical antibody to prevent antigens from invading the submucosal. You want IgA against H1N1. These plasma cells line the epithelial which secrete the dimeric IgA. The epithelial cell makes a recepter that binds to both IgA and transport it through the epithelial cell. When it gets to the lumen, the receptor will be cleaved and part of it remains associated with the IgA and becomes the secretory component. It can't be really cleaved due to the nature of its environment. IgE is present at low concentrations for allergies and is protective against parasites. IgD is primarily a membrane immunoglobulin. Even knockdown people appear normal. It is also interspecies so we just don't know why. All isotypes can exist both secreted (humoral) or membrane where they act as a receptor . Secreted are hydrophilic and membrane bound are hydrophilic to associate with the membrane. These are all weak forces, they only work a short distance. That means you have tight binding between the Ab binding site and the antigen. The Ab binding site needs to be able to close upon the Ag to get high affinity binding. The antigen is connected to the H and L chain. You see a aa that is protruding into the binding pocket. if you mutate this aa, the ab can't bind well. C, is a picture turned on it's side to see where it was recognized. You can see there are intereactions between two interacting surfaces. If you look at which aa are contacting, most of these interactions occur between the complementary regions. You have a lot of surface for the ab and the ag. These forces are weak, so you need a lot of tight binding. Epitope, the portion of the ag that is recognized, they are exposed on the surface of the protein. Different from T cells that recognize a cut up peptide of the foreign bacteria. Whereas, this present a native piece to be recognized. The protein is folded so that the aa that is recognized by the ab come together. they dont have to be together in linear sequence. The most epitopes are formed from non-linear that come together in 3D space. It's only a portion of the surface. On a protein antigen, there are many epitopes that can be recognized. ...
View Full Document
- Fall '06