Vous êtes sur la page 1sur 19

Basic immunology in 20 minutes

Joseph Barillari (this part) John McPhie (cool applications – stay tuned)

4 November 2004

Facts and graphics unashamedly stolen from Janeway, Travers, et al. Immunobiology, 6 th ed.; Hofmeyr. “An Overview of the Immune System”; Forrest, Hofmeyr. “Immunology as information processing”; Pillai, Hendrickson, et al. HST 175 course notes., Wikipedia.org, jpl.nasa.gov, etc.

Review slide – this is all in the paper.

What is the immune system?

Biological mechanism for identifying and destroying pathogens within a larger organism.

Pathogens: agents that cause disease

Bacteria, viruses, fungi, worms, etc.

Review slide – this is all in the paper.

Why would a computer scientist study the human immune system?

Massively parallel information processing mechanism – with 6.5B users!

Incredibly effective example of a distributed system built from diverse components which are constantly being renewed.

May inspire better computer security systems (stay tuned for next week), as it's

adaptive – can train self to react to new threats

error-tolerant – small mistakes are not fatal

self-protecting – protects itself

Roadmap

Many introductions present IS as a giant parts list.

We'll briefly consider it as a set of barriers, from a pathogen's perspective.

Disadvantage: lots and lots and lots of supporting details omitted. Ask questions.

Advantages: fast, tailored to hobby-horse concepts of immunologically-inspired security.

Namely negative selection, costimulation, combinatorial & junctional diversity, and somatic

hypermutation. (Don't ask questions about this line. We'll get to this.)

First line of defense is the innate immune system.

(This wasn't in Hofmeyr.)

Innate immunity

Pathogens

Skin

(ubiquitous) (think firewall) Hard-coded detectors for common pathogenic signatures. Macrophage Very few i ngests
(ubiquitous)
(think firewall)
Hard-coded detectors
for common
pathogenic signatures.
Macrophage
Very few
i
ngests & destroy
s
penetrate skin
m
ost pathogens
Very, very few
evade innate
immunity
Skin repels
nearly everything
Not to scale.

Rarely, nasty germs can evade the innate immune system. Enter adaptive immunity.

Adaptive immunity

T cells destroy infected cells to eradicate intracellular pathogens. (Some bacteria, all viruses)

B cells secrete antibodies to attack extracellular pathogens (Most bacteria)

T T B Bacterium T B Virus-infected cell
T
T
B
Bacterium
T
B
Virus-infected cell

The colors of the receptors indicate specificity: each can bind to one specific antigen. Adaptive immunity can only attack targets that it has prepared for.

Suppose you're a pathogen.

You've avoided the innate immune system.

Should the adaptive immune system give you pause?

Why would one of these receptors be able to bind to you?

Suppose you're new. As-yet-unseen.

Impossible, right?

T cell development, briefly outlined

Prototype cell, moves from bone marrow to thymus gland

T
T
Creates T cell receptor by sloppy gene rearrangement. (next slide) T
Creates T cell
receptor by
sloppy gene
rearrangement.
(next slide)
T
T
T
T
T
by sloppy gene rearrangement. (next slide) T T T T T Self-targeted T cells are deleted
T T
T
T

Self-targeted T cells are deleted (2 slides from now)

T T T Self-targeted T cells are deleted (2 slides from now) T Useful, non-self-reactive T
T
T
T T Self-targeted T cells are deleted (2 slides from now) T Useful, non-self-reactive T cells
T T Self-targeted T cells are deleted (2 slides from now) T Useful, non-self-reactive T cells

Useful, non-self-reactive T cells are released from the thymus

T cell receptor (TCR) creation

(1) TCR = V segment + D segment + J segment. Genome contains several different copies of each. Pick one from each set. (Combinatorial diversity.)

(2) Join the copies together using a sloppy technique that introduces randomness into the junctions. (Junctional diversity.)

(3) If you get an in-frame protein, continue.

V1

V4

V7

V2 V3 V5 V6 V8 V9 V6 V6 D4 D4
V2
V3
V5
V6
V8
V9
V6
V6
D4
D4
D1 D2 J1 J2 J3 J4 J5 J6 D3 D4
D1
D2
J1
J2
J3
J4
J5
J6
D3
D4

J1 J1

From a few dozen (to several dozen) segements for each category (V,D,&J) the human immune system creates over 10 11 different antibody receptors.

Janeway 136

binding strength

T cell refinement

Candidate T cells are exposed to most of the proteins in the human body. For presentation, proteins are chopped into fragments and displayed on specialized presentation molecules.

T T
T
T
T T
T T

Strongly autoreactive T cells are killed (negative selection)

autoreactive T cells are killed (negative selection) T cells with moderate binding strength are retained. T

T cells with moderate binding strength are retained.

T cells with moderate binding strength are retained. T T T cells from the green zone
T cells with moderate binding strength are retained. T T T cells from the green zone
T T
T T

T cells from the green zone are approved

are retained. T T T cells from the green zone are approved # of self peptides
are retained. T T T cells from the green zone are approved # of self peptides
are retained. T T T cells from the green zone are approved # of self peptides

# of self peptides bound

T cells that do not bind to any presentation molecules are allowed to die.

Summary of the last three slides

Your immune system has a repritoire of T cells capable of binding and destroying cells that exhibit almost any foreign protein.

Suppose you're a pathogen.

You've avoided the innate immune system.

Should the adaptive immune system give you pause?

Why would one of these receptors be able to bind to you?

Suppose you're new. As-yet-unseen.

Impossible, right?

Wrong!

It is unlikely that any intracellular-pathogen-infected cell will escape binding by a T cell.

The first binding will lead to replication of the bound T cell and (if all goes well) eventual clearance of the infection.

OK, so I'm toast. But what happens if a T cell binds a human cell? (They're moderately autoreactive, right?) Will the human cell be lysed?

if a T cell binds a human cell? (They're moderately autoreactive, right?) Will the human cell

Vern Virus

Answer: Costimulation

To be activated, a T cell needs to see both its first signal (the target antigen) and a second “DANGER” signal.

The “DANGER” signal is provided by antigen presenting cells when they detect signs of infection (cell lysates, certain cytokines, etc.)

T
T
signs of infection (cell lysates, certain cytokines, etc.) T Second (DANGER) First T cell signal receptors

Second

(DANGER)

First T cell signal receptors

T cells cannot be activated without a “DANGER” signal. If a T cell receives its first signal without the second, it may become tolerized to its target.

If you haven't noticed, bold italic means important!

I'm an extracellular pathogen. What should I expect?

I'm an extracellular pathogen. What should I expect? P i e r c e P a

Pierce Pathogen

Most pathogens are not self-aware.

B cells improve themselves via somatic hypermutation

B
B
B cells improve themselves via somatic hypermutation B Activation via B/T collaboration Clonal expansion B B

Activation via B/T collaboration

Clonal expansion

B B B
B
B
B
B Activation via B/T collaboration Clonal expansion B B B B Somatic hypermutation (random mutation of
B Activation via B/T collaboration Clonal expansion B B B B Somatic hypermutation (random mutation of
B Activation via B/T collaboration Clonal expansion B B B B Somatic hypermutation (random mutation of
B
B
Activation via B/T collaboration Clonal expansion B B B B Somatic hypermutation (random mutation of BCR)

Somatic hypermutation (random mutation of BCR)

B B B B Somatic hypermutation (random mutation of BCR) Advantageous mutations increase secreted antibody binding
B B B B Somatic hypermutation (random mutation of BCR) Advantageous mutations increase secreted antibody binding

Advantageous mutations increase secreted antibody binding affinity

B B
B
B

Deleterious

mutations