start immuno 1/6/97-----------------------

Dr.Farrell 
Immunoglobulins and antigens

B cells, antibody structure, Ig isotypes, how Ig works

Immunoglobulin is a globular protein produced by B cells which hangs out 
in the serum and ccan interact with foreign antigenic molecules. Ig is 
made of 2 heavy (larger) chains, which are identical, and two light 
chains which are also identical. These are held together by disulfide 
bonds. The N terminal portion is the Ig receptor, which can interact and 
bind to Ag molecules. This binding is an important effector mechanism in 
immunological response. 

in the body are millions of B cells - when they mature, they undergo 
immunoglobulin gene rearrangement which leads to the production of genes 
coding for the Ig, so each B cell produces a unique Ig molecule which is 
on its surface. during an immune response, antigenic molecules will 
interact with a few B cells, which happen to have the right Ag. so the 
basis of an immune response is for Ag to bind to specific B cells and 
induce them to produce antibody. After a short lag period is an increase 
in the Ab in the serum which will plateau and then decrease. if animal 
exposed to same Ag again, there is a asecondary, bigger response, where 
more Ab is made for longer. Because there is MEMORY in the immune system.


So if you look at selection of B cells you see that you start with a few 
B cells - an antigen binds to ones that have the right Ig, stimulating it 
(clonal selection) so they divide, increase in number, and become plasma 
cells which secrete Ab molecules. some of these B cells, though, will 
become memory cells, and will stay in reserve to recognize the antigen at 
a later date.

WHAT IS AN ANTIGEN

immunogenic molecules can elicit an antibody response when innoculated 
into the body

antigens can interact with antibody

not all antigens are immunogenic, but all immunogens are antigenic

haptens are substances which can combine specifically with Ab, but can't 
induce production of Ab - used experimentally - small antigens, 
basically.

properties of immunogens:

foreignness- "non self"
molecular size - the larger the molecule, the more immunogenic. protein 
Ag >5000-6000 good antigen; smaller eg < 1000 bad antigen. 

chemical nature
proteins usually strong Ag
carbohydrates: relatively weak, but when bound to glycoprotein, highly 
antigenic
lipids, usually not immunogenic, but can be, rarely
nucleic acid: rarely immunogenic. thought of as non-immunogenic. only see 
this in maybe autoimmune dz.

the more chemically complex the molecule, the more immunogenic.

old research - 30-40 yrs ago - characterizing Ig specificity.
these studies derived from landsteiner's work. he took protein molecules 

fig 14-3. he took an antigenic protein (A) which could be modified by 
substitution of small chemical moieties. he added dinitrophenol and then 
immunized an animal with the derivative protein, to see if Ab were made 
to the Dnp molecule. answer: yes, there were Ab made against the Dnp. 
then he manipulated the moieties of the protein - added aminobenzene: 
found Ab against it. when he substituted various carboxyl groups onto the 
aminobenzene, he found that the Ab elicited originally no longer reacted 
with the aminobenzene. when he immunized with derived proteins, he found 
that Ab against the original protein did not interact with the derived 
protein, and vice versa. so he proved the very highly specific nature of 
antibody. he did similar studies for a while, substituting lots of groups 
etc, and kept finding the same thing. see also tables 10.6 and 10.8 (all 
in handout)

bottom line: Ab for a specific Ag is very highly specific for that 
molecule. changing the molecule will affect ability of Ab to bind.

reason: Ig binding site must fit very closely with antigenic molecule for 
binding to occur. binding is not an irreversible event. Ag can bind Ab 
and in many cases releasing and rebinding may occur. the strength of the 
binding == the affinity, strength of Ab/Ag interaction. if there is a 
poor fit, the Ag/Ab complex is more likely to dissociate.

all binding bet Ag and Ab is noncovalent. these are very weak forces 
occuring over very short distance, eg H bonds, hydrophobic bonds, 
electrostatic bonds, van der waals bonds.

if we look at the Ab binding site, we see that the antigen fits into a 
groove formed by the heavy and light chains. contact points need to occur 
throughout that binding region. the closer the fit the greater the 
affinity.

ideally, one antigen will elicit one antibody which has very high 
affinity for it. this isn't what actually happens all the time. a variety 
of Igs have varying ability to bind to any particular antigen - so it's 
not an all or nothing thing. Ab porduced in response to 
paraaminophenolalphaglucoside interacts with that molecule, and also Ab 
made in response to the betaglucoside will interact with the 
alphaglucoside, just not as well. and vice versa. so some Ab have a high 
capacity to interact with a specific Ag, but others have a lesser 
capacity to interact with a specific Ag. an immune response will produce 
a number of Ig, some of which are highly specific for the Ag and others 
less specific.

PRODUCTION OF AB FOLLOWING IMMUNIZATION

if we look at a B cell, mature B cells have on their surfaces Ig 
molecules. the portion away from the membrane is the binding site (eg 
each molecule has two sites). think in terms of population. there are 
millions of B cells. what happens if we immunize with a particular Ag? if 
we give Ag A, that Ag has capacity to bind to B cells which express the 
correct receptor. there may be many B cells capable of binding A to 
various degrees. so on a population level, you may have a B cell that can 
bind A but can also bind Ag C and D which are closely related to A.  
there may be another B cell which is different, that can bind to A but 
then also can bind E and F and not C and D. So any one B cell can respond 
to a particular Ag and produce a specific Ab against it. but on a 
population level there may be multiple b cells with slightly altered 
specificity that all bind A, so we find a variety of closely related Ab 
being produced, which may all have separate slightly different 
specificities.

say a B cell has a receptor which can bind to Ag A. there may be another 
B cell with a receptor that is different but can still bind A. all b 
cells which can interact with A will be activated at immunization, but 
they will all produce sl different Ab. so you get a polyclonal response, 
generating a population of different Ig molecules.....etc.

if we look at a population level we often find that Ag used to elicit Ab 
response elicit both highly specific Ab and also Ab that are of slightly 
different specificity, that will also react with other Ag, leading to 
CROSS REACTIVITY phenomenon.

briefly: carbohydrates. basically not highly immunogenic unless bound to 
proteins. usually found as glycoproteins. the outer core of salmonella 
has a sugar chain -the O5 chain. if you look at different species of 
salmonella you see that Ab can be made to a glycoprotein present in all 
salmonella species. there are also antibodies that can be made against 
only one species. 

proteins. immunoglobulins bind to conformational determinants (?) not to 
linear portions of a molecule. see handout whale and chicken diagram. if 
a mouse or dog were immunized with the sperm whale myoglobin, Ab would be 
elicited that react to that myo w, and in a large molecule like that 
myoglobin, multiple Ab would be elicited b/c has multiple antigenic 
sites. say 4 Abs are elicited. 

the same thing would happen if you immunize a mouse with chicken eggwhite 
lysozyme. note in handout that two regions of the lysozyme when apposed 
by tertiary structure, become antigenic region. so antigens aren't just 
linear structures, but you have to look at the shape of the molecule 
which, if disrupted, will change the antigenicity of the molecule.

if you immunize animal with lysozyme, Ab are made against it.
if you excise out the loop peptide portion, the Ab will still recognize 
that portion. If you CLEAVE the loop, and make it a liinear structure, 
the Ab no longer reacts. If you make an Ab to the loop only and then 
cleave it, same thing, but that Ab will react with the native protein.

so tertiary structure is important in antigenicity.

now. (we'll come back to this later)

we discussed what Ab see 
during a normal immune response B cells don't react with Ag alone and 
make specific Ab. they cooperate with T cells which make multiple 
products (interleukins) to induce proliferation and Ab production and 
secretion.
immunogenic molecules are different from those with elicit T cell 
response.

look at Ig supergene family. you find antibody molecules attached to cell 
surface of B cells,and you find a similar structure called the T cell 
receptor on the T cell. Each T cell has a unique T cell receptor.so in 
the body you have millions of T cells, all specific to different 
antigens. Alsoon the cell surface are MHC molecules (class I or II.) 
Class I present on all cells, class II only on APCs which bind antigenic 
molecules and present them to T cells.

so. (see handout for T cell recep diagram)
T cells recognize peptides: linear portions of the peptide. they don't 
see DNA, lipid, CHO. they recognize small peptides. so the T cell 
receptor has a specific ability to interact with a peptide, but only when 
it is complexed to MHC II molecule.

so if you have an Ag, say a complex protein antigen, that antigen can be 
bound by a specific APC, which degrades it and binds it to MCH II and 
transports the peptide/MHC II complex to the surface and presents it to 
the T cells.

re: MHC 
in mouse chromosome 17.
a series of genes encoding for a series of molecules expressed on cell 
surfaces. some of them are class I molecules and some are class II. these 
define the histocompatibility between individuals. differences in these 
class I molecules are the basis of rejected transplants, etc. Class II 
molecules are ONLY present on APCs. these class II molecules are 
basically cell membrane proteins which can bind small linear peptide 
molecules. in terms of class I and II molecules class I can bind peptides 
8-12 AA s in length, class II binds 10-20 AA peptides. So an Ag is broken 
down to peptides and complexed to MCH II and exteriorized and presented 
to T cell.

class II MHC peptide interaction is not highly specific. MHC II can bind 
a broad range of linear peptide molecules. can't bind ALL of them, but 
many. so during this degradation of the antigen, SOME of the peptides 
will bind MHC II but not all will.

so, T cell receptor unlike B cell recep recognizes PROCESSED Ag bound to 
MHC II. so the complex stimulates the T cell receptor; native antigen 
does NOT.

In any one given protein there will be some peptides capable of binding 
MHC II. Myelin basic protein causes an immunological response...mice of 
two haplotypes expressing two different MHC types bind two different 
peptides within the MBP molecule. in two different rats, same situation- 
each rat binds a different peptide. but each animal has a response. so 
different portions of the MBP are immunogenic depending on the MHC of the 
particular host animal and the capacity to bind peptides.

if you take a myoglobin peptide you can show that it will activate a 
specific T cell. you can start shortening the peptide and maintain 
reactivity but when you shorten it too much you eliminate the reactivity, 
so you can determine the specific sequence needed to produce the 
reactivity. minimum portion about 10 AAs. linear portions of peptides are 
immunogenic to T cells. conformational determinants of whole native 
antigen are immunogenic to B cells.

----break----

:)  Hi!

ok, let's get started again.
immunology is kind of a circular subject. you need to hear about things 
out of context and then come back to them later and stuff. we'll try to 
talk more about MHC later.

Immunoglobulin structure. see front page of handout.
two identical heavy chains flanked by two identical light chains. the 
light chains have two varieties. two different constant regions either 
kappa or lambda. heavy chains have many types: M, G, D, E, A...IgG uses a 
gamma chain, IgE an epsilon chain, etc.

the light chains are linked to heavy chains by disulfide bonds. within 
the chains are many internal disulfide bonds. there are also disulfide 
bonds w/in heavy chains. the N terminal end is the Ig binding site aka 
variable region. so for an IgG molecule you have two identical Ig binding 
sites, one made by each pair of chains. within the Ig molecule are also 
constant regions: constant region of light chains and of heavy chains. 
these regions are highly conserved between molecules. the variable 
regions are highly variable. their variability leads to the production of 
different binding sites and hence specificity.

there are carbohydrates on the Ab molecule. right in front of disulfide 
bonds joining the two chains is the hinge region. that allows for some 
flexibility of the molecule at that site. so the Ab binding site can 
actually move to better interact with the Ag, so if Ab is binding to 
surface of bacteria, the binding sites can be closer or farther apart as 
needed. Constant region has three parts all encoded for by different 
genes. if you look at the variable region you find that the linear 
sequence of AAs is unique on each light and heavy chain for each molecule 
capable of binding a distinct Ag. within these variable regions are 
regions called hypervariable regions. these are different to an even 
greater extent between Ig molecules. that is because these regions form 
the actual Ig binding site.so if you look at amino acid variability 
across the light chains of many Ig molecules you see in some regions a 
very high region of AA varability - these are the complementarity 
determining regions or something like that. he's muttering. ok, he just 
repeated it, i had it correct :) those regions are basically the 
hypervariable regions: same thing. between these regions are the 
framework regions which are less variable but still more variable than 
the constant region which is basically identical between molecules.

so if you look at the structure of an Ig molecule, at the Ab binding site 
(Fab) it's the tertiary structure which will determine the binding site. 
since this is a globular protein, you see that the hypervariable regions 
are exposed at the terminus of the molecule to be part of the binding 
site. and the links between the light and heavy chains are such that the 
hypervariable regions of light and heavy chains come together at the end 
of the molecule to form a hypervariable cleft with very high specificity 
for a particular antigen.

[note: bottom line so far: each immunoglobulin has a unique Ag 
specificity.]

now if you look at electrophoresed serum, you find different molecules eg 
albumin, Ig fraction (IgM, IgG, IgE) these are all called immunoglobulin 
isomers. if you look at an IgG molecule it has the structure we've been 
discussing; often drawn as a Y.

\\ //
  ||
  ||

but there are multiple types eg IgG1 IgG2 IgG3 IgG4. these differ mainly 
due to the constant regions of the heavy chains which differ between Ig 
TYPES and these changes affect the biological properties.

also have IgM molecules which are pentamers in secreted form. Basically 
composed of 5 Ig molecules linked together by a joining chain. looks like 
5 IgG molecules all linked at the Fc region (spokes of wheel 
arrangement). IgA (secretory Ab of saliva, GI, lung) is a dimer of two Ig 
molecules...looks like two Ys joined at the Fc regions joined by a J 
chain and with a "secretory component"wrapped around it, which is 
produced by the secretory cell, which allows the molecule to be secreted 
into the extracellular space.

finally IgE, associated w/allergy,  looks to me like IgG.

how do we get all these forms?
there are multiple genes encoding the different portions of these 
molecules.
quickly we'll discuss gene rearrangement. all cells in the body have the 
same germ line something. so all cells have DNA capable of encoding a 
variety of different speciificities of Ig molecules. but these cells must 
be ACTIVATED.

As B cells mature in the bone marrow from stem cells, they go through 
multiple stages...proBcells, etc. during that time, gene rearrangement 
occurs in the Ig variable region genes for light and heavy chains. the 
molecules are made of variety of different regions. during development 
from stem to early proB cell a rearranging process occurs. one specific 
variable region becomes aligned with an adjoining region

eg, need to choose one of several Fv genes, and one of several "adjoining 
region" genes, and that happens, and they are put with the Fc gene. the 
same thing happens with the heavy chain. there is a huge variety of 
variable region genes, and adjoining (J) region genes, and also there are 
some D region genes. during maturation one of each is linked together to 
form the variable region of the heavy chain. 

the constant region in the light chain must be one of two types, kappa or 
lambda. it's more complex with the heavy chain. there is a series of 
constant region genes

mu encodes for IgM
delta IgD
gamma IgG
epsilon IgE
alpha IgA

as the B cells develop they will first express on cell surface a 
rearranged variable region gene for light and heavy chains and individual 
constant region gene. so in early B cell youfirst express IgM. then 
before fully mature you start expressing IgD. Mature B cells express IgM 
and IgD on their surfaces. the antibody specificity is locked into that 
cell because the variable region genes have been rearranged already. but 
the heavy chain constant region can be changed still. so a B cell as a 
naive B cell will have a given Ig specificity and will express monomeric 
IgM and IgD on the surface of the cell. during an immune response a B 
cell can switch heavy chain it uses while maintaining specificity, so it 
can make different Ab isotypes, altering functional activity of Ab.

variable regions - there are about 40 genes for kappa and 29 for lambda, 
51 genes for variable heavy chain regions.

it's the interaction of all these genes which allows for this specificity 
and the wide range of uniquely specific antibodies which are produced. 
the specificity is locked into the cell once it is mature, but it can 
still make differnt TYPES of antibody, all against the same antigen - by 
altering the heavy chain constant region gene being expressed.

note: see handout; much of this is diagrammed within it.

studies on Ig structure led to nobel prize back in 50s or so. showed that 
Ig molecules can be cleaved by papain or pepsin. pepsin cleaves behind 
disulfide bonds, leaving the Fab regions separate from the Fc fragments. 
if papain is used, it cleaves anterior to disulfide bonds, leaving two 
Fab portions and an Fc portion. studies showed that different parts of 
the molecule have different capacity to interact with other things. 
showed that Ag binding site is in the variable region - Vh and Vl 
(variable heavy and ligh). the Cgamma1 portion of heavy chain binds 
complement C4b fragment. the Cgamma2 portion binds C1q complement, 
controls catabolic rxns.

the Fc receptor portion binds to mononuclear cells, eos, neuts, plates, 
is involved in opsonization, I can't see the stupid slide from here and 
he's not reading it so oh well.

affinity and avidity. the capacity of an individual Ab binding site is 
discussed as antibody affinity. the whole molecule's capacity to bind Ag 
is discussed as valence - an IgG has a valence of two: has two binding 
sites. IgM has a valence of 10 (if all 10 sites bind antigen.)

antibody variants
in an individual animal all constant regions are identical - isotypic. in 
an outbred population there are slight differences in constant regions of 
the kappa chains but they are functionally identical - allotypic - 
follows mendelian rules of heredity. idiotypic variation refers to the 
unique amino acid sequences present within the Ab binding site. in 
theory, each Ab molecule has a unique structure and is different from 
every other Ab molecule.


unique variable region gene encodes binding site of each Ig molecule. for 
something to be antigenic it has to be foreign. this holds true for all 
molecules int he body except Ig molecules. since each Ig is unique, it 
also has the capacity to elicit an antibody against it, and this has been 
shown to be the case. during an immune response where you secrete a lot 
of Ab and the body is flooded with a specific Ab, other B cells may 
interact with the unique tertiary structure of the FIRST antibody and 
make Ab against it!

outside the Ab binding site and within the Ab binding site are unique 
areas.
so an antibody may BIND to the unique receptor portion of another 
antibody, or to an antigenic region upon it, or something. theoretically, 
and Ab that binds with high specificity to the binding site of another 
Ab, would be the mirror image of the Ag originally intended to bind to 
that site. those antigens then can elicit antibodies against self. some 
will interact outside the binding site or on the binding site. these 
antibodies are called anti-idiotypic Ab. there are also 
anti-anti-idiotypic Abs. Jerne's network hypothesis illustrates a 
mechanism by which all Ab may interact with other Ab during immune 
response and can help to dampen or amplify an immune repsonse by 
mimicking antigen or antibody. in fact if one inoculates an Ag and looks 
at Ab that is produced (the idiotypic Ab), you see that shortly after 
that, you get anti-idiotypic Ab.

we know this can occur in theory but we aren't sure how they actually 
function to regulate an immune response.some vaccines were attempted with 
anti-antibodies.


----end-----