Human preimmune repertoire needs a
capacity of ~ 1015 to cope with antigens
encountered in life time.
Human genome has ~ 105 genes, not enough
to code for all in preimmune repertoire.
Antibody Diversity
Clonal selection
•– Each B cell makes only one antibody and is a
clone (each cell has ~105 immunoglobulin
molecules on the surface).
•The binding of antigen to the receptor on the surface
of naïve cell (when the cell is in the peripheral
lymphoid organ)activates the B cell clone to develop
into effector cells and to proliferate
Clonal selection
Clonal selection
After the first exposure, some become memory
cells.
Both memory and naïve B cells are activated by
antigen binding.
Memory cells respond differently than naïve cells,
resulting in a higher antibody titer
Clonal selection
Immunoglobulins or Antibodies
• Classes of antibodies (differ in their heavy chain)
• 5 classes of heavy chain IgA, IgD, IgE, IgG, IgM with α, δ,
ε, γ, μ heavy chain respectively IgG, IgA have subclasses
(IgG1, IgG3, IgG3, IgG4 with γ1, γ2, γ3, γ4 respectively).
• 2 light chain κ or λ
• Constant region and variable region
• Both heavy chain and light chain have a constant region
and a variable region.
•Three hypervariable section are the loci which are really
variable in the variable region of heavy and light chains,
the rest of “framework” in the variable region is relatively
invariant.
The antibody molecule
Antibody diversity is generated by
a number of mechanisms
• Combinational recombination
Mammal recombine immunoglobulin gene segments before
transcription.
•Each chain (λ, κ, and heavy chains) has a pool of gene segments.
• Each pool is on a different chromosome. i.e. on one chromosome.
• There are a large pool of segments for variable region and a
smaller pool of constant region.
• During B cell development, a coding sequence is assembled by site
specific recombination to increase the diversity. An antibody IgD
with a К light chain thus has 7 x 106 possible combination.
Each variable region of a light chain is made
of a V and a J gene segment
Human heavy chain gene segment
pool
•Each variable region of a heavy chain is
madeof a, V, J, D segment
Imprecise V(D)J joining
•Only appropriate segments get joined (V always
joins to J in a light chain, not V to V). It is site
specific.
• Catalysed by RAG-1 and RAG-2 (recombination
activating genes)
• During recombination, nucleotides can be lost or
inserted (may shift reading frame)
• Can have a second round of V(D)J rearrangement
(receptor editing)
Mouse immunoglobulin genes
V1-V500D1-D12 J1-J4
H chain locus (Chr 12)
CCC3C1
C2bC2aCC
V1-V250 J1-J5 C
V2 J2C2
chain locus (Chr 6)
chain locus (Chr 16)
C3 C1V1 J3 J1
Mechanisms for generating
antibody diversity
V(D)J recombination
Somatic hypermutation
Class switch recombination
V1-V500 D1-D12 J1-J4 constant regions
V(D)J recombination
Mechanisms for generating
antibody diversity
V1-V500 D1-D12 J1-J4 constant regions
Hozumi and Tonegawa, PNAS 1976
V(D)J recombination
Mechanisms for generating
antibody diversity
V1-V500 D1-D12 J1-J4 constant regions
V(D)J recombination
Mechanisms for generating
antibody diversity
V1-V500 D1-D12 J1-J4 constant regions
V(D)J recombination
D to J joining
Mechanisms for generating
antibody diversity
V(D)J recombination
D to J joining
Mechanisms for generating
antibody diversity
V(D)J recombination
V to DJ joining
Mechanisms for generating
antibody diversity
** *
V(D)J recombination
Somatic hypermutation
Class switch recombination
Class switch
Mechanisms for generating
antibody diversity
V(D)J recombination
Somatic hypermutation
Class switch recombination
** *
Class switch
Mechanisms for generating
antibody diversity
V(D)J recombination
Somatic hypermutation
Class switch recombination
** *
V(D)J recombination and class switch recombination
involve double-strand breaks
Mechanisms for generating
antibody diversity
Recombination signal sequence (RSS)
direct V(D)J recombination
23
CACAGTG–––––––ACAAAAACC
23
GGTTTTTGT–––CACTGTG
12
heptamer nonamernonamer heptamer
12
V
J
The 12/23 rule: recombination occurs between
a RSS with a 12 bp spacer and a RSS with a 23 bp spacer
V(D)J recombination involves DNA
cleavage and end-joining
coding joinsignal join
cleavage
end-joining
Cleavage is initiated by RAG1/RAG2
(recombination activating genes)
RAG1/RAG2 nick DNA at two RSSs....
then catalyze nucleophilic attack by 3' OH on the opposite strand
5' OH
5' OH
5'
5'
generating hairpin coding ends and blunt signal ends
van Gent, Gellert et al. Cell 1995
DNA ends are modified by
addition and deletion
N-nucleotide addition by terminal
deoxynucleotidyl transferase (TdT)
P-nucleotide addition by asymmetric
opening of hairpin coding ends
Nucleotide deletion
Addition at DNA ends
TCA
AGT
TCA
AGT
N-nucleotide addition
addition at 3' ends by TdT
addition
Germ-line gene Rearranged gene Mechanism
GATCA
CTAGT
P-nucleotide addition
hairpin formation and
asymmetric cleavage
addition
TCA
AGT
GATCA
T
ATAGT
TATCA
Deletion at DNA ends
.
Germ-line gene Rearranged gene Mechanism
CA
GT
TCA
AGT
Nucleotide deletion
random
deletion
TCA
AGT
Nucleotide deletion
microhomology directed
deletion
TCA
AGT
CA
GT
T
C
A
G
TC
A
G
T
DNA pathways in V(D)J recombination
Evolution of V(D)J recombination
RAG1 and RAG2 contain no introns and
are tightly linked on the same chromosome
RAG1 and RAG2 are conserved back to
the evolution of jawed fish
Evolutionary hypothesis: a transposon
with RAG1, RAG2, and associated RSSs
infected a precursor of jawed fish
RAG1 and RAG2 do not exist
in jawless fish
hagfishlamprey
Origin of the immunoglobulin genes
D JV
ancient receptor gene
V
gene duplication
D12 J4V500 C
first transposon integration
second transposon integration
Mechanisms for repairing
DNA double-strand breaks
Homologous recombination
Non-homologous end-joining
V(D)J recombination mutants are defective
in non-homologous end-joining
Proteins involved in
non-homologous end-joining
Protein Enzymatic activity
Ku DNA end-binding
DNA-PKcs DNA-dependent protein kinase
XRCC4/ Ligase 4 DNA ligase
Artemis exonuclease
Rad50/ Mre11/ Nbs1 exo/endonuclease
Somatic hypermutation (SHM)
SHM targets immunoglobulin genes
(but not T cell receptor genes)
SHM requires active transcription
SHM involves DNA single-strand breaks
Model for somatic hypermutation
Activation-induced deaminase (AID)
Expressed only in activated B cells
Converts C to U in single-stranded DNA
Other proteins insert mutations
Uracil DNA glycosylase converts U to an apurinic site
AP endonuclease nicks the DNA adjacent to the AP site
Exonuclease removes the AP ribose
An error-prone polymerase fills in the gap
Model for somatic hypermutation
How is C mutated on both strands with the same frequency?
How does SHM target the Ig locus, but not other loci?
Class switch recombination (CSR)
CSR rearranges the constant regions to
generate different antibody isotypes
CSR regions
located 5’ to each C
H gene, except for C
consist of repeats of GAGCT and GGGGGT;
e.g., switch region is [(GAGCT)
nGGGGGT]
150
CSR requires active transcription
AID initiates CSR
CSR occurs via double-strand breaks
CSR requires Ku and DNA-PKcs
CSR junctions show characteristics of
non-homologous end-joining
Deletions to regions of microhomology
Duplications from DNA polymerase activity
Model for class switch recombination
How does AID initiate CSR at one locus and SHM at another?
(The C-terminus of AID is required for CSR, but not SHM.)