----start immuno 1.10.97----


John Wolfe
Handout: lecture 4- the genetic and cellular basis of the antibody 
repertoire.


we're going to talk about how antibodies get made biologically by immune 
cells. cell biology and genetic mechanisms that create this enormous 
repertoire of exquisite specificity. this large globular molecule does 
have an exquisite specificity and this is what forms the basis of the 
ability of the immune system to distinguish foreign proteins.it's 
*highly* specific. you must NOT be immune to the hundreds of thousands of 
SELF protein, yet must BE immune to many other unknown proteins.

today: genetics, cell biology. how do B cells get to the point where they 
can make Ab?

there are about 100,000 genes in the human body, give or take 50k. We can 
make 10000000 or 100000000 antibodies, though. three genes govern this.

[slide- basic IgG structure] there are hypervariable regions in Fab 
region which have a unique tertiary structure which will only fit a 
specific antigen.

keep in mind that the heavy and light chains are polypeptides. they are 
coded in order, of course, determined by RNA. it's how the RNA is made 
that results in the coding for the hypervariable region, which then folds 
into the appropriate tertiary/quaternary structure to form the Ab binding 
site. so it's all based on structure, which is determined by sequence, 
which is determined by genes. most genes have at most a few variants. an 
individual inherits a specific variant. but THIS variability occurs on 
the individual level.

the Ab binding site is very small. the AAs in the binding site determine 
the shape of the the binding site- lock and key mechanism. it's very 
small and that is why it has such exquisite specificity. changing ONE AA 
will change the specficity.

V region variability:  (see handout). there are hypervariable regions 
within the variable regions of the light and heavy chains.
why  do we call these hypervariable? in a single molecule it isn't 
variable, but it is variable between molecules of antibody. you have 
antibodies with related specificity, but not quite the same.

immunoglobulin gene superfamily. 
Igs have small cytoplasmic region, then globular extracellular region 
after transmembrane sequence. The T cell receptors are similar, with 
similar structure, and are as specific as B cell Igs- both of these have 
variable regions. class I and II MCH molecules also have similar 
structures. these are the "transplantation antigens" and they do not have 
a variable region - they vary on the population level. NCAM is a neural 
cell adhesion molecule which also has a similar structure but no variable 
region (i think no variable region)
but there are a bunch of molecules with very similar structures, and the 
variable region is unique to T cell receptors and immunoglobulins, is the 
bottom line. all come from differet genes.

it's very important to understand this next part to understand how three 
genes can generate this specificity. we should have had some of this 
before.
need to understand CLONALITY and EXPANSION FROM SUBSET of cells.

pluripotent stem cell: can reconstitute entire blood cell line
        |
 proliferation followed by differentiation
        /             \
lymphoid lineage    myeloid lineage
    |                   |
B and T cells      granulocytes
                   erythrocytes
                  monocytes
                  megakaryocytes
   

the proliferation and differentiation is governed by genes. proliferation 
can be enormous- geometric expansion. so you have a stem cell which 
follows a series of steps to differentate into a uncommited precursors 
and then finally COMMITTED precursors, eg myelod and lymphoid stem cells. 
what makes them committed? well, the stem cells are exposed to 
interleukins which are one factor which drives them down one pathway or 
the other. many labs are working to increase our understanding. but 
differentiation isn't INHERENT in the cell. the stem cells respond to 
signals from the environment to go one way or the other, or to 
proliferate simply. stem cells are also a self renewing population. when 
they divide, one daughter stays a stem cell, and one becomes committed. 
some of the stem cells become lymphoid precursors, and then T and B 
cells, after responding to different signals which drive them down the 
different pathways. [note: this is in the handout and the book as well.]

note that there are also highly specific functional subsets of B and T 
cells.

you can't tell them apart histologically, but you can tell them apart by 
their genes. there are genetic B cell differentiation markers. note that 
some of these "markers" are relevant to the function of the cell type, 
others we dunno.
the preB cell is marked by Cµ; the plasma cell is marked by clg, TIO. 
these are surface proteins that are useful to determine subtypes of 
cells.

the basic structure of genes and how they are made is necessary to 
understand because Ig genes are an exception.

you need an enhancer element upstream to tell the gene to be on.

enhanc.---//----promoter---*|AUG|--|  |--| exon |-intron-|  |---|    TAA 
|AAAAAAAA----

at dna level this is just a string of codons. the RNA is transcribed from 
the *start site tothe poly A site in a base for base process.

sequence of Heavy chain gene rearrangements. the Hchain locus has to be 
put into form such that it will work. before the stem cell gets the right 
signals, eg when in germ line configuration, the cell can't make Ig. the 
B cell rearranges this locus. this is only a property of 7 genes, all in 
B and T cells, and it only works when it is in the pre B or preT cell 
stage. hepatocytes, for example, have these genes, but cannot express 
them because they cannot rearrange them correctly. so in the B cells, you 
can have a productive rearrangement, which will work, so the cell goes on 
to be productive, or you can have a rearrangement that doesn't have the 
capacity to make the heavy chain, it doesn't line up right -so it will 
try the other allele. if it messes up again, it will probably die, 
whatever it does it isn't going to make Ig. Note that only ONE allele 
will productively rearrange, not both.

germline Ig genes
see handout. bottom right determines entire constant region (CH part)
v=variable
d=diversity
j=joining
all of these are segments of DNA too small to code a protein. the gene 
rearranges to put the segments next to each other and then hooks it all 
up with that constant part. so you have hundreds of segments put 
togeteher, and you need a heavy chain and a light chain. the combinations 
of putting a hundred of these together...you have a large number of 
possible combinations. what's the range of the large number? 
10000000-100000000.

ultimately- a small segment that encodes a specific segment of a small 
chain of polypeptides something something argh. he's confusing me.

in an individual cell coming from a single clone, once this rearrangement 
occurs the clonal progeny will have that rearrangement and will make that 
peptide. this is the source of monoclonal antibodies.

the GERM line config is what is inherited in all the cells, but the 
properties of B cells (and T I suppose) allow for the rearrangement.

for heavy chain, this is how it works....see handout again.
this large stretch of DNA undergoes rearrangement such that one variable 
region, one diversity region and one joining region are brought together 
IN FRAME (otherwise it's non productive) - eg it has to be spliced 
together correctly or you won't get protein production. as far as we know 
this is a totally random process that preB cells undertake. it does NOT 
happen in other cells. notice when the DNA rearrangement occurs, segments 
of DNA are actually LOST. spliced out. bye bye. and the cell can NOT 
revert to the original genome. cell division is occuring at the same 
time, see.

once this occurs, a basic genetic change has been made. this process 
occurs first, before the lightchain rearrangement, which is similar, but 
it happens to occur in the heavy chain first.

so you get transcription int o nuclear mRNA, then splicing, and you get 
cytoplasmic mRNA

this will determine the structure of the Ig mu heavy chain. it has a 
genetic sequence that will be translated into an AA sequence just like in 
any regular cell. it's already been rearranged. other cells have 
different sequences since they had other rearrangements. but of course 
the CONSTANT region is the same in all of the cells, because it does't 
rearrange. basically once this is done, the cell has made a unique gene 
segment. 

Ig splice recognition sequence...don't have to write this down (ha) but 
the idea is to put these pieces together. the gene has these flanking 
sequences-= of 23 bases, specific to this locus, and also a 12 base 
sequence on the other end--that have complementarity to each other and 
can recombine. these segments come together to bring two other parts  in 
proximity, and then are spliced out in a very unique mechanism to put 
these pieces of DNA together. but the DNA itself is also designed to be 
recognized by this splicingmechanism that is used. don't confuse this 
with RNA splicing- different thing.
-------break------

this unique rearranging mechanism has other genes involved in making it 
work, and some of those are recombinase (an enzyme poorly understood) - a 
very unusual enzyme, because DNA splicing doesn't usually occur in vivo. 
the cell does have this enzyme though, and other genes as well...probably 
the recombinase gets turned on by recombinase activating gene - rag - so 
the rag gene has to be turned on first to activate the enzyme and then 
the enzyme makes the mechanism work. and there are probably a lot of 
other things along this line also. if you make KO mice transgenic for a 
rag deletion, you knock out the immune system. it has to have this to 
function normally. now normally you wouldn't want dna recombinase 
floating around in your cells, but this enzyme is highly specific for VDJ 
joining (Just VJ in light chains of course)

now, to reiterate, the light chain is a similar process to the above 
delineated heavy chain process and occurs subsequent to it. you create a 
unique nucleic acid sequence encoding a unique AA combo to create the 
light chain. in the kappa thereis one constant region, lambda has a 
couple of constant regions.

realize this all occurs before the cell has ever seen an antigen.

there is also splicing of the mRNA within this locus of course. the mRNA 
is spliced into an in frame coding sequence. we start with the VDJ joined 
already of course from before. the - what is shown in the diagram is that 
in this locus where you have mu and delta you get transcripts which 
include the single VDJ segment. so it gets transcribed, and spliced into 
a single segment which has aspecific nucleic acid sequence....through 
differential splicing,the whole VDJ locus is transcribed, but different 
splicing puts together the delta heavy chain and the mu heavy chain, both 
with different constant regions and the same VDJ region. this will be 
DIFFERENT when we talk about IgG. (we're apparently discussing IgM now 
*shrug*). so this is called differential splicing. the function and 
mechanism is poorly understood but it occurs, and helps cells to modulate 
their use of gene segments. from this transcript you can make two 
different copies but the variable part is the same and constant region 
differs.

PROCESS
germ line form/stem cell: unrearranged form. antigen independent phase
VDJ rearrangement of µ chain: light chain not rearranged yet
preB cell with cytoplasmic µ chains
VJ rearrangement of L chain: immature B cell: has heavy and light 
chains,IgM
Mature B cell: has IgM and IgD---final step of Ag independent phase
          |---start Ag dependent phase----
        antigen and helper factors encourage proliferation and 
differentiation
          |
          |
         / \
        /   \
Plasma cell  memory cells with surface Ig: they can make the same Ig for 
the                same Ag when it is encountered again.

understand that the once the heavy and light chains are expressed and you 
have IgM and IgD on the surface

splicing for membrane or secreted IgM
membrane bound form has AA sequence which is hydrophobic, so is retained 
in membrane, and secreted form has different AA sequence in that spot. 
this is differential splicing as before - RNA processing, not DNA 
rearrangement.

diversity generation:

chain combination: H+L, a+b
combinational: VDJ
truncation of D or J
N segment
Junctional
---above all in Ig and T cell recep--

somatic hypermutation: in Ig only: for the most part, those cells which 
best fit the antigen will undergo expansion, so there is kind of a 
natural selection for those cells which make Ig which is matched with the 
Ag.
as these cells grow and expand, since body exposed to mutagens all the 
time, these cells do have mutations, and these loci are subject to 
mutagenesis. 
if the mutation changes the codon into a different aa, and that aa is in 
the Ab binding pocket, it will change the binding specificity, and 
perhaps it will have BETTER binding to the Ag of interest, then the 
daughter cells will be favored for expansion as well...so the effect you 
see when you measure Ab in serum is the sum of the contribution of 
thousands of B cells, and so this is called "maturing of the response" - 
and you see an increased binding to Ag over time. but genetic 
rearrangement is the MAIN generator of diversity, andthis is not the 
major way.

three reading frames: T cell recep only.


In the handout: b cell development and clonal selection. note the various 
B cells of different specificity. the left half is the Ag independent 
phase. in the middle you see an Ag with three epitopes, and the different 
B cells respond to different epitopes on the Ag. on the right is the 
resultant expansion. now, if you have a local response in one LN you will 
only have the B cells there responding. in the Ag dependent phase, you 
may have cells with related specificity not recognize this Ag. realize 
that the naive B cell can only respond to an Ag which FITS its Ig 
receptor. won't respond to just any old Ag.

so you get cell expansion, through only a few cell divisions. this 
results in what you see in the serum. in a primary immune response, you 
see an initial lag phase after introduction of Ag and then you see 
polyclonal IgM response because the cell has IgM on the surface. shortly 
after that you see an IgG response (class switching). then you have a 
decline, first IgM then IgG. in a SECONDARY response you see a similar 
IgM response after a MUCH bigger, faster, longer lasting IgG response. - 
see handout. to get the switch from IgM to IgG requires another set of 
genetic events. in the simplest form, Ag processing of peptides occurs, 
presented to B cells, and T cells, each recognizing specific part, and 
they cooperate with presenting cells, helper cells, respond to cytokines, 
and this makes B cell become plasma cell, which is required to make IgG.

switch regions facilitate class switching:
VDJ stays the same, but cell undergoes a second DNA level gene 
rearrangement, losing a chunk of DNA, and you the locus rearranged. you 
need to notice this is very important to specificity. you keep the same 
VDJ, same antigen specificity, but you switch from making IgM (membrane 
bound Ag) to IgG. after dna rearrangement, the normal rna processing 
occurs. so you have an IgG constant portion fused to the same VDJ 
variable region.
you can have two successive class switches- IgM to IgG to IgE (assoc 
w/allergies, etc.) so you can get the same specificity hooked up to 
different constant region segments to give it different functions. but 
you can't return to the original class - you lose DNA each time.

---end----